Information processing device, information processing method, and information processing program

Abstract

Content related to the evaluation subject will be provided according to the impact that the added elements have on the evaluation of the evaluation subject. [Solution] The information processing device according to the present invention is characterized by having an estimation unit that estimates an additional object whose influence on the evaluation of the evaluation subject by a predetermined user satisfies predetermined conditions, based on a first evaluation which shows an evaluation from an evaluator for first target information which is information that shows only the evaluation subject, and a second evaluation which shows an evaluation from an evaluator for second target information which shows additional objects along with the evaluation subject, and a provision unit that provides content related to the evaluation subject to the user based on the additional objects estimated by the estimation unit.

Description

Technical Field

[0004] , , ,

[0006] , , , ,

[0005] , , ,

[0007] , , ,

[0001] The present invention relates to an information processing apparatus, an information processing method, and an information processing program.

Background Art

[0002] Conventionally, a technique for collecting evaluations of an evaluation target by evaluators and analyzing the evaluators is known. As an example of such a technique, a technique for analyzing the sensitivity of users based on the evaluation information of users with respect to an evaluation target (such as a design) has been provided.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] However, with the above-described technique, it is not always possible to provide content related to the evaluation target according to the influence that the target added to the evaluation target has on the evaluation of the evaluation target (the sensitivity of the user).

[0005] For example, with the above-described technique, it only outputs a design image based on the analyzed sensitivity of the user, and it is not always possible to provide content related to the evaluation target according to the influence that the target added to the evaluation target has on the evaluation of the evaluation target.

[0006] The present application has been made in view of the above, and an object thereof is to provide content related to an evaluation target according to the influence that the target added to the evaluation target has on the evaluation of the evaluation target.

Means for Solving the Problems

[0007] The information processing device according to the present application is characterized by comprising: an estimation unit that estimates an additional object whose influence on the evaluation of the evaluation subject by a predetermined user satisfies predetermined conditions, based on a first evaluation which shows an evaluation from an evaluator for first target information which is information that shows only the evaluation subject; a second evaluation which shows an evaluation from an evaluator for second target information which shows additional objects along with the evaluation subject; and a provision unit that provides content related to the evaluation subject to the user based on the additional objects estimated by the estimation unit. [Effects of the Invention]

[0008] According to one embodiment, the effect is achieved that content related to the evaluation target can be provided in accordance with the influence that the object attached to the evaluation target has on the evaluation of the evaluation target. [Brief explanation of the drawing]

[0009] [Figure 1] Figure 1 shows an example of the configuration of the information processing system 1 according to an embodiment. [Figure 2] Figure 2 shows an example of information processing according to the present invention. [Figure 3] Figure 3 is Figure (1) showing an example of target information according to the embodiment. [Figure 4] Figure 4 is Figure (2) showing an example of target information according to the embodiment. [Figure 5] Figure 5 shows an example of the configuration of the information processing device 10 according to the embodiment. [Figure 6] Figure 6 shows an example of the target information database 31 according to the embodiment. [Figure 7] Figure 7 shows an example of the evaluator information database 32 according to the embodiment. [Figure 8] Figure 8 shows an example of a user information database 33 according to the embodiment. [Figure 9] Figure 9 is a flowchart (1) showing an example of the information processing procedure according to the embodiment. [Figure 10]Figure 10 is a flowchart (2) showing an example of the information processing procedure according to the embodiment. [Figure 11] Figure 11 is a flowchart (3) showing an example of the information processing procedure according to the embodiment. [Figure 12] Figure 12 is a hardware configuration diagram showing an example of a computer that implements the functions of the information processing device 10. [Modes for carrying out the invention]

[0010] The following describes in detail, with reference to the drawings, the embodiments for implementing the information processing device, information processing method, and information processing program according to the present application (hereinafter referred to as "embodiments"). Note that these embodiments do not limit the information processing device, information processing method, and information processing program according to the present application. Furthermore, the same parts are denoted by the same reference numerals in each of the following embodiments, and redundant descriptions are omitted.

[0011] (Embodiment) [1. Configuration of the Information Processing System] First, an information processing system 1 according to an embodiment will be described. Figure 1 is a diagram showing an example configuration of the information processing system 1 according to an embodiment. As shown in Figure 1, the information processing system 1 includes an information processing device 10, a user terminal 100, and an evaluator terminal 200. The information processing device 10, the user terminal 100, and the evaluator terminal 200 are connected to each other via a predetermined communication network (network N) by wired or wireless means. Note that the information processing system 1 shown in Figure 1 may include multiple information processing devices 10, multiple user terminals 100, and multiple evaluator terminals 200.

[0012] The information processing device 10 is an information processing device that receives evaluations from an evaluator regarding target information that indicates the subject of evaluation, and realizes information processing according to the evaluation, and is realized by, for example, a server device or a cloud system. For example, in the example shown in Figure 2, the information processing device 10 receives evaluations from an evaluator (annotator) regarding target information that indicates the subject of evaluation, which is a combination (also called a coordinate or outfit) of multiple clothing items (also called fashion items, including footwear (also called shoes), hats (e.g., caps, hats, etc.), accessories (also called accessories), and small items (bags, etc.)), indicating whether the combination is favorable or not (in other words, whether the combination of clothing is appropriate or not, whether the combination of clothing is compatible or not, whether the combination of multiple clothing items looks good or not), and realizes information processing according to the evaluation. Furthermore, the information processing device 10 may receive an evaluation from an evaluator of target information that represents a single garment as the subject of evaluation, indicating whether the single garment is favorable or not (in other words, whether the garment is appropriate or not, whether the garment is good or not, or whether the garment suits the person or not), and then perform information processing according to the evaluation. In addition, the evaluation can be from various perspectives. For example, the information processing device 10 may receive an evaluation from an evaluator indicating whether a combination of multiple garments (or a single garment) is cool or not, or whether a combination of multiple garments (or a single garment) is cute or not, or whether a combination of multiple garments (or a single garment) is fashionable or not, or whether a combination of multiple garments (or a single garment) is in style or not, and is not limited to these examples. Thus, the content of the evaluation can be any, depending on the situation.

[0013] Furthermore, for example, the information processing device 10 provides an e-commerce service that offers clothing (sale, rental, etc.). The information processing device 10 also provides a coordination service that accepts submissions from users of content (for example, still images, videos, articles, etc.) that shows clothing coordination and provides it to other users.

[0014] Incidentally, the information processing apparatus 10 may have a function as a web server that provides a website related to an e-commerce service or a coordination service. Further, the information processing apparatus 10 may be a device that distributes information to be displayed on an application related to an e-commerce service or a coordination service, which is installed on the user terminal 100 or the evaluator terminal 200, to the information processing apparatus 10. Further, the information processing apparatus 10 may be a device that distributes the application data itself.

[0015] Further, the information processing apparatus 10 may function as a distribution device that distributes control information to the user terminal 100 or the evaluator terminal 200. Here, the control information is described by, for example, a script language such as JavaScript (registered trademark) or a style sheet language such as CSS (Cascading Style Sheets). Note that the application itself distributed from the information processing apparatus 10 may be regarded as control information.

[0016] The user terminal 100 is an information processing apparatus used by a user. The user terminal 100 is realized by, for example, a smartphone, a tablet-type terminal, a notebook PC (Personal Computer), a desktop PC, a mobile phone, a PDA (Personal Digital Assistant), or the like. Further, the user terminal 100 displays information distributed by the information processing apparatus 10, a server apparatus that provides a predetermined service, or the like, on a web browser or an application. Note that in the example shown in FIG. 2, the case where the user terminal 100 is a smartphone is shown.

[0017] The evaluator terminal 200 is an information processing device used by evaluators to evaluate the target information. The evaluator terminal 200 can be implemented as, for example, a smartphone, tablet, notebook PC, desktop PC, mobile phone, or PDA. The evaluator terminal 200 displays information distributed by the information processing device 10 and server devices that provide predetermined services, using a web browser or application. In the example shown in Figure 2, the evaluator terminal 200 is a smartphone.

[0018] [2. An example of information processing] Next, an example of information processing realized by the information processing device, etc. according to this embodiment will be described using Figure 2. Figure 2 is a diagram showing an example of information processing according to the embodiment. In the following description, it will be assumed that the user terminal 100 is used by a user (user U1) identified by the user ID "UID#1". Also, in the following description, the user terminal 100 may be treated as the same as user U1. That is, in the following, user U1 can be read as user terminal 100.

[0019] Furthermore, in the following explanation, evaluator terminals 200-1 to 200-N (where N is any natural number) will be used depending on the evaluator using evaluator terminal 200. For example, evaluator terminal 200-1 is evaluator terminal 200 used by evaluator A1, identified by evaluator ID "AID#1". Also, in the following explanation, evaluator terminals 200-1 to 200-N will be referred to simply as evaluator terminal 200 without any particular distinction. In addition, in the following explanation, evaluator terminal 200 may be treated as the same as the evaluator. That is, in the following explanation, the evaluator can be read as evaluator terminal 200.

[0020] Furthermore, the following explanation will show an example of classifying multiple evaluators with corresponding attributes into groups G1 to G4 and presenting target information to evaluators belonging to each group. Here, each evaluator belonging to groups G1 to G4 must correspond to at least one of the following attributes: gender, age, place of residence or workplace, preferred fashion categories, preferred fashion brands, preferred fashion genres, preferred fashion influencers, degree of interest in fashion (fashion sensitivity) (for example, "low" indicating no interest, "medium" indicating some interest, or "high" indicating a high level of interest, which may be estimated from search trends, browsing trends, purchasing trends, or purchase amounts), sources of fashion inspiration (homepages, social media, magazines, etc. that are frequently accessed or bookmarked), purchasing trends, purchase amounts (monthly, yearly), or occasions in which they frequently wear fashion (for example, frequently attend weddings). Furthermore, multiple evaluators may be classified into each group regardless of their attributes (for example, based on the order in which they logged in or randomly).

[0021] Furthermore, in the following explanation, outfits #1 through #25 will be the ones being evaluated.

[0022] Furthermore, in the following explanation, the information processing device 10 presents two pieces of information to the evaluator and accepts an evaluation indicating which piece of information shows a favorable code (in other words, which piece of information wins when the two pieces of information compete against each other). The information processing device 10 then repeats this process and determines a rating (Elo rating) for each piece of information based on each evaluation.

[0023] First, the information processing device 10 selects target information that indicates the evaluation target to be presented to the evaluator (step S1). For example, the information processing device 10 selects two pieces of target information to present to evaluators belonging to groups G1 to G4 from a group of target information #1, which consists of multiple pieces of target information corresponding to each of the outfits #1 to #25, and which are images that show only outfits #1 to #25 (in other words, images that show only the clothing, without including information such as the head (face (before or after makeup) and / or hairstyle) or body (body shape) of the person wearing the clothing that makes up the outfit) (in other words, target information without additional elements).

[0024] To give a specific example, when selecting target information to present to evaluator A1 from target information group #1, the information processing device 10 first selects target information #1 from target information group #1 that currently has the fewest number of evaluations (in other words, number of matches) indicating that it has been evaluated by each evaluator belonging to group G1. Then, based on the ratings of each target information included in target information group #1 (ratings based on the evaluation results of each evaluator belonging to group G1), the information processing device 10 selects target information whose rating difference with target information #1 is within a predetermined range (in other words, target information that is estimated to have a high probability of being a draw with target information #1). To give one example, the information processing device 10 selects target information #2 that has the smallest rating difference with target information #1 (in other words, target information #2 that is estimated to have the highest probability of being a draw with target information #1).

[0025] In this way, by matching information that is likely to result in a draw (in other words, information that makes it difficult to determine which outfit is more favorable), the amount of information obtained through the evaluator's assessment can be increased.

[0026] Next, the information processing device 10 presents the target information to the evaluator (step S2). For example, the information processing device 10 presents target information #1 and target information #2 to evaluator A1. To give a specific example, the information processing device 10 displays target information #1 and target information #2 side by side on the screen of evaluator terminal 200-1, allowing evaluator A1 to select whichever one they rate favorably.

[0027] Next, the information processing device 10 receives an evaluation of the presented target information from the evaluator (step S3). For example, the information processing device 10 receives an evaluation from evaluator A1 indicating which of target information #1 and target information #2 is more favorable. The information processing device 10 then adds 1 to the evaluation counts of target information #1 and target information #2 in group G1, and sets the ratings for target information #1 and target information #2 in group G1 based on the evaluation from evaluator A1.

[0028] For example, if the information processing device 10 evaluates target information #1 as more favorable, the smaller the difference in ratings between target information #1 and #2, the higher the value that increases the rating of target information #1 and the higher the value that decreases the rating of target information #2. Also, if the information processing device 10 evaluates target information #1 as more favorable and target information #1 has a higher rating than target information #2, the larger the difference in ratings, the lower the value that increases the rating of target information #1 and the lower the value that decreases the rating of target information #2. Also, if the information processing device 10 evaluates target information #1 as more favorable and target information #2 has a higher rating than target information #1, the larger the difference in ratings, the higher the value that increases the rating of target information #1 and the higher the value that decreases the rating of target information #2.

[0029] Furthermore, the information processing device 10 performs the same processing as in steps S1 to S3 above when it receives further evaluations of the target information from evaluator A1, or when it receives evaluations of the target information from other evaluators. Here, after the information processing device 10 receives evaluations of target information #1 and target information #2 from evaluator A1, when selecting target information to present to evaluators belonging to group G1, the information processing device 10 selects target information according to the number of evaluations of each target information in group G1, which has changed according to the evaluation, and the rating of each target information in group G1. In other words, the information processing device 10 does not present the same combination of target information to each evaluator, but rather selects a combination of target information according to the number of evaluations of each target information, which changes from time to time, and the rating of each target information, and presents it to the evaluator.

[0030] Next, the information processing device 10 determines whether or not to exclude the evaluator's evaluation results based on the evaluator's evaluation of the target information (step S4). For example, if evaluator A2 always or more than a predetermined percentage of the same evaluation (for example, if they always or more than a predetermined percentage of the target information displayed in a predetermined area (for example, the right or left area) on the screen of the evaluator terminal 200), or if they evaluate in a certain pattern (for example, if they select the target information displayed in an area on the screen of the evaluator terminal 200 in a certain pattern (for example, alternately)), or if the time from when the target information is displayed to when it is selected is greater than or equal to a predetermined threshold (if the time until evaluation is too long), or less than or equal to a predetermined threshold (if the time until evaluation is too short), etc. (in other words, if the evaluator is presumed to be unserious about evaluating the target information), the information processing device 10 excludes the evaluator's evaluation results from the evaluation results in group G1, and further sets evaluator A2 not to be a target for presenting the target information (in other words, sets them as an excluded person). Furthermore, evaluator A2 may be moved to group G5 (a group of evaluators presumed to be unserious), where they may be asked to select and present the target information and continue their evaluation. If, within that group, the evaluator is presumed to be serious about evaluating the target information (in other words, not presumed to be unserious about evaluating the target information), they may be returned to their original group, where they may be asked to select and present the target information and resume their evaluation. Additionally, evaluator A2 may not be an unserious evaluator, but rather an evaluator with specific preferences. For example, they may appear to be giving an unserious evaluation precisely because they have specific preferences. Therefore, evaluator A2 may be moved to group G6 (a group of evaluators presumed to have specific preferences), where they may be asked to select and present the target information and continue their evaluation. Based on the evaluation results within that group, the ranking within that group may be determined.Furthermore, as will be described later, if the information processing device 10 presents dummy information along with (or in place of) the target information in the original group or group G6, and the evaluator evaluates the dummy information favorably, then the evaluator is not an evaluator with specific preferences, and therefore the evaluator's evaluation results may be excluded.

[0031] Next, the information processing device 10 excludes the evaluation results of excluded individuals and selects the target information to be presented to evaluators who are not designated as excluded individuals (step S5). For example, the information processing device 10 recalculates the number of evaluations for each target information in group G1 by subtracting the number of evaluations by evaluator A2. The information processing device 10 also excludes the evaluation results of evaluator A2 and recalculates the rating for each target information in group G1. Then, based on the recalculated number of evaluations and ratings for each target information, the information processing device 10 selects the target information that indicates the evaluation target to be presented to evaluators belonging to group G1. Note that because the information processing device 10 subtracts the number of evaluations by evaluator A2, target information with a reduced number of evaluations is more likely to be selected (in other words, target information with a reduced number of evaluations is preferentially displayed on the screens of 200 evaluators belonging to group G1 other than evaluator A2).

[0032] Next, the information processing device 10 presents the selected target information to evaluators who have not been designated as excluded (step S6) and accepts their evaluation of the presented target information (step S7). Then, based on the evaluation content of the target information, the information processing device 10 determines whether or not to exclude the evaluator's evaluation result (step S8). Note that the processing in steps S6 to S8 is the same as in steps S2 to S4, so the explanation is omitted.

[0033] The information processing device 10 performs the above processing until the number of evaluations performed by each evaluator for the target information belonging to target information group #1 exceeds a predetermined threshold (for example, 25 times or more).

[0034] The information processing device 10 may calculate the accuracy of the rating of the target information belonging to target information group #1 in each group, and perform the above processing until the accuracy is above a predetermined threshold. For example, the more times target information #1 in group G1 has been evaluated, the higher the accuracy of the rating of target information #1 in group G1 will be calculated by the information processing device 10, and the fewer times target information #1 in group G1 has been evaluated, the lower the accuracy of the rating of target information #1 in group G1 will be calculated.

[0035] Furthermore, if an excluded person is designated, the information processing device 10 may designate a new evaluator, present the target information to that evaluator, and accept the evaluation.

[0036] After the above processing for target information group #1 is completed, the information processing device 10 selects two pieces of target information to present to evaluators belonging to groups G1 and G3 from target information group #2, which consists of multiple pieces of target information corresponding to each of the outfits #1 to #25, and which further show images of the heads of people wearing the outfits in addition to the images of the outfits (in other words, target information to which the additional target "head" has been added to the outfits). The information processing device 10 then presents the selected target information and accepts evaluations of the presented target information. The information processing device 10 also determines whether or not to exclude the evaluation results from the evaluators based on the content of the evaluations given to the evaluators. These processes are carried out until the number of evaluations by each evaluator for the target information belonging to target information group #2 exceeds a predetermined threshold. Note that these processes are the same as in steps S1 to S8, so the explanation is omitted.

[0037] On the other hand, after the above processing for target information group #1 is completed, the information processing device 10 selects two pieces of target information to present to evaluators belonging to groups G2 and G4 from target information group #3, which consists of multiple pieces of target information corresponding to each of the outfits #1 to #25, and which are images showing the body shape of a person wearing outfits #1 to #25 (in other words, images of the person wearing the outfit from the head down (images that do not include the head)) (in other words, target information to which the attached target "body shape" has been attached to the outfit). The information processing device 10 then presents the selected target information and accepts evaluations of the presented target information. The information processing device 10 also determines whether or not to exclude the evaluation results by the evaluators based on the content of the evaluations given to the evaluators. These processes are carried out until the number of evaluations by each evaluator for the target information belonging to target information group #3 exceeds a predetermined threshold. Note that these processes are the same as in steps S1 to S8, so the explanation is omitted.

[0038] Furthermore, after the above processing for target information group #2 is completed, the information processing device 10 selects two pieces of target information to present to evaluators belonging to groups G1 and G3 from target information group #4, which consists of multiple pieces of target information corresponding to each of the outfits #1 to #25, and which are images showing a person wearing outfits #1 to #25 (in other words, images showing the head and body shape of a person wearing the outfits) (in other words, target information to which the attached elements "head" and "body shape" have been attached to the outfits). The information processing device 10 then presents the selected target information and accepts evaluations of the presented target information. The information processing device 10 also determines whether or not to exclude the evaluation results from the evaluators based on the evaluation content from the evaluators. These processes are carried out until the number of evaluations by each evaluator for the target information belonging to target information group #4 exceeds a predetermined threshold. Note that these processes are the same as in steps S1 to S8, so the explanation is omitted.

[0039] Furthermore, after the above processing for target information group #3 is completed, the information processing device 10 selects two pieces of target information from target information group #4 to present to evaluators belonging to groups G2 and G4. The information processing device 10 then presents the selected target information and accepts evaluations of the presented target information. The information processing device 10 also determines whether or not to exclude the evaluator's evaluation results based on the evaluation content of the evaluator. These processes are carried out until the number of evaluations by each evaluator for the target information belonging to target information group #4 exceeds a predetermined threshold. Note that these processes are the same as steps S1 to S8, so the explanation is omitted.

[0040] In other words, the information processing device 10 presents target information from target information groups #1, #2, and #4 to groups G1 and G3, and presents target information from target information groups #1, #3, and #4 to groups G2 and G4. By presenting target information selected from at least one similar target information group to multiple groups in this way, for example, if the groups are classified according to the attributes of the evaluators, it is possible to determine whether similar evaluation results (ratings) can be obtained between group G1 and group G3 (in other words, whether evaluation results (ratings) can be obtained for each attribute, whether an evaluation result (rating) can be obtained by integrating the evaluation results (ratings) of group G1 and group G3, or whether the evaluation results (ratings) are converging). Furthermore, if the groups are not classified according to the attributes of the evaluators, it is possible to determine whether similar evaluation results (ratings) can be obtained between Group G1 and Group G3 (in other words, whether there is a bias in the attributes of the evaluators (if there is a bias in the attributes of the evaluators, should the groups be moved), whether an integrated evaluation result (rating) can be obtained by combining the evaluation results (ratings) of Group G1 and Group G3, and whether the evaluation results (ratings) are converging).

[0041] Here, we will explain examples of target information to be presented to the evaluator using Figures 3 and 4. Figure 3 is Figure (1) showing an example of target information according to the embodiment. Figure 4 is Figure (2) showing an example of target information according to the embodiment.

[0042] As shown in Figure 3, the information processing device 10 first selects and presents two pieces of target information to evaluators belonging to groups G1 and G3 from target information group #1, which shows only the outfits, as shown in image C11, and accepts their evaluation. Subsequently, the information processing device 10 selects and presents two pieces of target information to evaluators belonging to groups G1 and G3 from target information group #2, which shows not only the outfits but also images of the heads of people wearing the outfits, as shown in image C12, and accepts their evaluation. Subsequently, the information processing device 10 selects and presents two pieces of target information to evaluators belonging to groups G1 and G3 from target information group #4, which shows images of people wearing the outfits, as shown in image C13, and accepts their evaluation.

[0043] Furthermore, as shown in Figure 4, the information processing device 10 first selects and presents two pieces of target information to the evaluators belonging to groups G2 and G4 from the target information group #1, which shows only the outfits, as shown in image C21, and accepts their evaluation. Subsequently, the information processing device 10 selects and presents two pieces of target information to the evaluators belonging to groups G2 and G4 from the target information group #3, which consists of images showing the body shape of a person wearing the outfit, as shown in image C22, and accepts their evaluation. Subsequently, the information processing device 10 selects and presents two pieces of target information to the evaluators belonging to groups G2 and G4 from the target information group #4, which consists of images showing a person wearing the outfit, as shown in image C23, and accepts their evaluation.

[0044] For example, if evaluator A1 evaluates target information showing outfit #1 with the attachment "head" attached, and then evaluates target information showing outfit #1 without the attachment, evaluator A1 may be unable to properly evaluate outfit #1 alone because the previously presented attachment "head" comes to mind.

[0045] Therefore, as described above, the information processing device 10 presents the target information to groups G1 and G3 in the order of target information groups #1, #2, and #4 (in other words, from the least amount of information to the most), and presents the target information to groups G2 and G4 in the order of target information groups #1, #3, and #4. This allows the information processing device 10 to obtain an appropriate evaluation of the target information.

[0046] Returning to Figure 2, the explanation continues. Next, the information processing device 10 estimates the influence of the added object on the evaluation of the outfit based on the relative relationships of the evaluators' evaluations of the target information groups #1 to #4 (step S9). For example, based on the rating of target information group #1 in group G1, the information processing device 10 determines ranking #1 for outfits #1 to #25 in a state where no added objects have been added. Also, based on the rating of target information group #2 in group G1, the information processing device 10 determines ranking #2 for outfits #1 to #25 in a state where the added object "head" has been added. Also, based on the rating of target information group #4 in group G1, the information processing device 10 determines ranking #3 for outfits #1 to #25 in a state where the added objects "head" and "body shape" have been added.

[0047] The information processing device 10 then uses ranking #3 as the ground truth data to calculate a score Sc1 indicating the degree of agreement between the ranks of outfits #1 to #25 in ranking #1 and the ranks of outfits #1 to #25 in ranking #3 (for example, a score indicated by a number ranging from "1" indicating a perfect match with ranking #3 to "0" indicating no match at all with ranking #3). The information processing device 10 also uses ranking #3 as the ground truth data to calculate a score Sc2 indicating the degree of agreement between the ranks of outfits #1 to #25 in ranking #2 and the ranks of outfits #1 to #25 in ranking #3.

[0048] Furthermore, the information processing device 10 determines ranking #4 for outfits #1 to #25 in a state where no additional items have been added, based on the rating of target information group #1 in group G2. Furthermore, the information processing device 10 determines ranking #5 for outfits #1 to #25 in a state where the additional item "body type" has been added, based on the rating of target information group #2 in group G1. Furthermore, the information processing device 10 determines ranking #6 for outfits #1 to #25 in a state where the additional items "head" and "body type" have been added, based on the rating of target information group #4 in group G1.

[0049] The information processing device 10 then uses ranking #6 as the ground truth data to calculate a score Sc3 indicating the degree of agreement between the ranks of outfits #1 to #25 in ranking #4 and the ranks of outfits #1 to #25 in ranking #6. The information processing device 10 also uses ranking #6 as the ground truth data to calculate a score Sc4 indicating the degree of agreement between the ranks of outfits #1 to #25 in ranking #5 and the ranks of outfits #1 to #25 in ranking #6.

[0050] Furthermore, the information processing device 10 determines ranking #7 for outfits #1 to #25 in a state where no additional items have been added, based on the rating of target information group #1 in group G3. Furthermore, the information processing device 10 determines ranking #8 for outfits #1 to #25 in a state where the additional item "head" has been added, based on the rating of target information group #2 in group G1. Furthermore, the information processing device 10 determines ranking #9 for outfits #1 to #25 in a state where the additional items "head" and "body type" have been added, based on the rating of target information group #4 in group G1.

[0051] The information processing device 10 then uses ranking #9 as the correct data to calculate a score Sc5 indicating the degree of agreement between the ranks of outfits #1 to #25 in ranking #7 and the ranks of outfits #1 to #25 in ranking #9. The information processing device 10 also uses ranking #9 as the correct data to calculate a score Sc6 indicating the degree of agreement between the ranks of outfits #1 to #25 in ranking #8 and the ranks of outfits #1 to #25 in ranking #9.

[0052] Furthermore, the information processing device 10 determines ranking #10 for outfits #1 to #25 in a state where no additional items have been added, based on the rating of target information group #1 in group G4. The information processing device 10 also determines ranking #11 for outfits #1 to #25 in a state where the additional item "body type" has been added, based on the rating of target information group #2 in group G1. Finally, the information processing device 10 determines ranking #12 for outfits #1 to #25 in a state where the additional items "head" and "body type" have been added, based on the rating of target information group #4 in group G1.

[0053] The information processing device 10 then uses ranking #12 as the ground truth data to calculate a score Sc7 indicating the degree of agreement between the rankings of outfits #1 to #25 in ranking #10 and the rankings of outfits #1 to #25 in ranking #12. The information processing device 10 also uses ranking #12 as the ground truth data to calculate a score Sc8 indicating the degree of agreement between the rankings of outfits #1 to #25 in ranking #11 and the rankings of outfits #1 to #25 in ranking #12.

[0054] Here, in the example in Figure 2, assume that scores Sc1 to Sc8 have been calculated as shown in graph Gr1. The information processing device 10 estimates the influence of the added element based on each score shown in graph Gr1. For example, in the example in Figure 2, assume that score Sc2 is higher than score Sc1 by a predetermined threshold, and score Sc6 is higher than score Sc5 by a predetermined threshold (in other words, the evaluation results are converging between group G1 and group G3). In such a case, the information processing device 10 estimates that the added element "head" has an influence on the evaluation of the code because when the added element "head" is added to code #1 to #25, the degree of agreement with the correct data changes by a predetermined threshold compared to when the added element "head" is not added. To give a specific example, the information processing device 10 estimates that when the item "head" is added to outfits #1 to #25, the degree of agreement with the correct data improves, and therefore the item "head" has a positive effect on the evaluation of the outfits (in other words, when the item "head" is added, the outfit is more likely to be evaluated favorably by the evaluator). In other words, the information processing device 10 estimates that it is important to consider the head of the person wearing the outfit when determining whether the outfit (the fashion items that make it up) is appropriate and whether the outfits are compatible. In other words, when the information processing device 10 recommends an appropriate or compatible outfit, it is preferable to consider the head of the person wearing the outfit.

[0055] Furthermore, in the example in Figure 2, score Sc4 is higher than score Sc3 by a predetermined threshold, and score Sc8 is higher than score Sc7 by a predetermined threshold (in other words, the evaluation results converge between group G2 and group G4). In such a case, the information processing device 10 estimates that when the "body type" attribute is added to outfits #1 to #25, the degree of agreement with the correct data changes by a predetermined threshold compared to when the "body type" attribute is not added, and therefore estimates that the "body type" attribute has an influence on the evaluation of the outfits. To give a specific example, the information processing device 10 estimates that when the "body type" attribute is added to outfits #1 to #25, the degree of agreement with the correct data improves, and therefore estimates that the "body type" attribute has a positive influence on the evaluation of the outfits (in other words, when the "body type" attribute is added, the outfits are more likely to be evaluated favorably by the evaluator). In other words, the information processing device 10 estimates that it is important to consider the body type of the person wearing the outfit when determining whether the outfit is appropriate or whether the outfit is a good match. In other words, when the information processing device 10 recommends an appropriate or compatible outfit, it is preferable that it takes into account the body type of the person who will wear the outfit.

[0056] Furthermore, in the example in Figure 2, scores Sc4 and Sc8 are higher than scores Sc2 and Sc6, so the information processing device 10 estimates that the added element "body shape" has a greater positive influence on the evaluation of the outfit than the added element "head." In other words, the information processing device 10 estimates that it is more important to consider the body shape of the person wearing the outfit than the head when determining whether the outfit is appropriate or whether the outfit is compatible. In other words, when recommending an appropriate or compatible outfit, it is preferable for the information processing device 10 to consider the body shape of the person wearing the outfit rather than the head when making recommendations.

[0057] Furthermore, if score Sc1 is higher than score Sc2 by a predetermined threshold, and score Sc5 is higher than score Sc6 by a predetermined threshold, the information processing device 10 may infer that adding the attachment "head" to outfits #1 to #25 results in a lower degree of agreement with the correct data compared to not adding the attachment "head," and therefore the attachment "head" has a negative impact on the evaluation of outfits (in other words, outfits with the attachment "head" are less likely to be rated favorably by evaluators). In other words, the information processing device 10 infers that it is important not to consider the head of the person wearing the outfit when determining whether an outfit is appropriate or whether the outfits are compatible. In other words, it is preferable for the information processing device 10 to recommend appropriate or compatible outfits without considering the head of the person wearing the outfit.

[0058] Furthermore, if score Sc3 is higher than score Sc4 by a predetermined threshold, and score Sc7 is higher than score Sc8 by a predetermined threshold, the information processing device 10 may infer that adding the "body type" attribute to outfits #1 to #25 results in a lower degree of agreement with the correct data compared to not adding the "body type" attribute, and therefore the "body type" attribute has a negative impact on the evaluation of outfits (in other words, outfits with the "body type" attribute are less likely to be evaluated favorably by evaluators). In other words, the information processing device 10 infers that it is important not to consider the body type of the person wearing the outfit when determining whether an outfit is appropriate or whether the outfit is compatible. In other words, it is preferable for the information processing device 10 to recommend appropriate or compatible outfits without considering the body type of the person wearing the outfit.

[0059] Furthermore, if scores Sc2 and Sc6 are higher than scores Sc4 and Sc8, the information processing device 10 estimates that the attached element "body shape" has a greater negative impact on the evaluation of the outfit than the attached element "head." In other words, the information processing device 10 estimates that considering the head is more important than the body shape of the person wearing the outfit when determining whether the outfit is appropriate or whether the outfit is compatible. In other words, when recommending an appropriate or compatible outfit, the information processing device 10 prefers to consider the head rather than the body shape of the person wearing the outfit.

[0060] Alternatively, a ranking for each group of target information could be determined based on the rating for each group of target information, and it could be estimated for each group whether each added element has a positive or negative impact on the evaluation of the outfit.

[0061] Next, the information processing device 10 generates new target information to present to the evaluator based on the influence that the added elements have on the outfit or fashion item. For example, the information processing device 10 generates new content related to the outfit or fashion item to which the added elements "head" and / or "body shape," which are estimated to have a positive influence on the evaluation of the outfit or fashion item, are attached, as new target information to present to the evaluator. To give a specific example, if the information processing device 10 estimates that the added elements "head" and "body shape" have a positive influence, it generates new images showing a person wearing the outfit or fashion item from images showing only the outfit or fashion item, as new target information to present to the evaluator. In this way, the information processing device 10 can increase the number of images showing a person wearing the outfit or fashion item, and determine a more accurate ranking of the outfit or fashion item with the added elements "head" and "body shape."

[0062] Next, the information processing device 10 provides the user with content related to the outfit or fashion item based on the influence that the added object has on the outfit or fashion item (step S10). For example, the information processing device 10 provides user U1 with content related to the outfit or fashion item to which the added object "head" and / or added object "body shape" that are estimated to have a positive influence on the evaluation of the outfit or fashion item have been added. To give a specific example, in an e-commerce service or a coordination service, the information processing device 10 provides content related to the outfit or fashion item to which the added object "head" and / or added object "body shape" have been added. To give a more specific example, when the information processing device 10 provides content showing an image of the outfit or fashion item, it provides content showing an image of the outfit or fashion item to which the added object "head" and / or added object "body shape" of user U1 or a person designated by user U1 (such as user U1's family, lover, friend, or gift recipient) has been added (for example, a head and body shape have been combined). Furthermore, the information processing device 10 provides content showing images of outfits or fashion items to which similar head and / or body type attributes are attached, as well as to user U1 or the target person. In this case, information (e.g., images or numerical information) about the head and body type of user U1 or the target person specified by user U1 is provided by user U1 or the target person specified by user U1. To give a more specific example, the information processing device 10 sets the priority of content related to outfits or fashion items to which head and / or body type attributes are attached higher than the priority of content related to outfits or fashion items to which head and / or body type attributes are not attached, and provides preferentially to content related to outfits or fashion items based on this set priority, or provides content showing their ranking.

[0063] Furthermore, for example, the information processing device 10 provides user U1 with content relating to outfits or fashion items that takes into account the additional elements "head" and / or "body shape" which are estimated to have a positive influence on the evaluation of the outfit or fashion item. To give a specific example, the information processing device 10 provides content relating to outfits or fashion items that take into account the additional elements "head" and / or "body shape" which are estimated to have a positive influence on the evaluation of the evaluation target, which is the outfit or fashion item, using a model that has been trained as training data, which is determined based on the ranking determined based on the evaluation results of the target information to which the additional elements "head" and / or "body shape" which are estimated to have a positive influence on the evaluation of the evaluation target. To give a more specific example, when the information processing device 10 receives input consisting of a first target group, which is a combination of some of the fashion items that make up the outfit to be evaluated; a second target group, which is a combination of the remaining parts of the fashion items that make up the outfit to be evaluated; and an additional target, which is "head" and / or "body shape" that is estimated to have a positive influence on the evaluation of the outfit to be evaluated, the information processing device 10 calculates a set matching score using a model that has been trained to output a higher set matching score the higher the degree of matching between the first target group, the second target group, and the additional target (the closer it is to a predetermined evaluation target with a higher ranking (a predetermined first target group and predetermined second target group that makes it up) and a predetermined additional target). The set matching score indicates the degree of matching between the first target group, the second target group, and the additional target to be calculated. Then, based on the calculated set matching score, the information processing device 10 provides content related to the combination of the first target group, the second target group, and the additional information to be calculated.For example, the information processing device 10 may provide content relating to outfits that are combinations of a first target group and a second target group that are subject to calculation, taking into account the additional target "head" and / or the additional target "body shape" that are subject to calculation, or content relating to multiple fashion items that constitute an outfit that is a combination of a first target group and a second target group that are subject to calculation, taking into account the additional target "head" and / or the additional target "body shape" that are subject to calculation. In addition, for example, the information processing device 10 may provide content relating to the additional target "head" and / or the additional target "body shape" that are subject to calculation, which are subject to calculation, and which match with the outfit that is a combination of the first target group and the second target group that are subject to calculation, or content relating to the degree of matching (set matching score) between the outfit that is a combination of the first target group and the second target group that are subject to calculation, and the additional information that is subject to calculation, which is the additional target "head" and / or the additional target "body shape" that are subject to calculation.

[0064] Furthermore, the target information used to determine the rankings for training the model may be prepared for each fashion genre or for each occasion. The information processing device 10 may determine the rankings for each fashion genre or occasion based on the evaluation results of the target information prepared for each fashion genre or occasion. This allows the model to learn rankings for each fashion genre or for each occasion, and the information processing device 10 can use such a model to provide content related to outfits or fashion items for each fashion genre or occasion.

[0065] Furthermore, if the content provided to user U1 does not include an attachment target "head" for the outfit or fashion item, the information processing device 10 will provide content in which the attachment target "head" has been added to the outfit or fashion item. To give a specific example, the information processing device 10 will provide content using model #1, which has been trained to generate an image of the outfit or fashion item with the attachment target "head" added from an image of the outfit or fashion item.

[0066] Furthermore, if the content provided to user U1 does not include the attached "body type" for the outfit or fashion item, the information processing device 10 will provide content in which the attached "body type" has been added to the outfit or fashion item. To give a specific example, the information processing device 10 will provide content using model #1, which has been trained to generate an image of the outfit or fashion item in which the attached "body type" has been added, from an image of the outfit or fashion item.

[0067] Here, Model #1 is trained to output a training image when it is input, for example, an image of only the outfit shown in the training image (an image of an outfit or fashion item with the attachment target "head" or "body type" attached). Any known technique can be applied to training Model #1, and a training method may be appropriately selected depending on the information used as training data. For example, Model #1 may be trained using various conventional machine learning techniques (e.g., supervised learning machine learning techniques such as SVM (Support Vector Machine)). Furthermore, deep learning techniques may be used to train Model #1. For example, various deep learning techniques such as RNN (Recurrent Neural Network) and CNN (Convolutional Neural Network) may be used to train Model #1.

[0068] Furthermore, the information processing device 10 may provide content related to outfits or fashion items to which the attached attributes "head" and / or "body type" have been attached, if user U1 or a person designated by user U1 has attributes corresponding to evaluators belonging to groups G1 to G4.

[0069] Furthermore, if it is estimated that the attached element has a negative impact on the evaluation of the outfit or fashion item, the information processing device 10 may provide content relating to the outfit or fashion item without the attached element. To give a specific example, the information processing device 10 provides content relating to the outfit or fashion item without the attached element in an e-commerce service or a coordination service. To give a more specific example, when the information processing device 10 provides content showing an image of an outfit or fashion item, it provides content showing an image of an outfit or fashion item without the attached element "head" and / or attached element "body shape" of user U1 or a person designated by user U1. Also, for example, if it is estimated that the attached element "head" has a negative impact on the evaluation of the outfit, the information processing device 10 sets the priority of content relating to outfits or fashion items with the attached element "head" lower than the priority of content relating to outfits or fashion items without the attached element "head," and provides preferentially content relating to outfits or fashion items based on the priority set in this way, or provides content showing their ranking. Furthermore, if the content provided to user U1 includes an outfit or fashion item with an attached "head," the information processing device 10 provides content with the attached "head" removed from the outfit or fashion item.

[0070] Furthermore, for example, the information processing device 10 provides user U1 with content relating to outfits or fashion items in which the attached elements "head" and / or "body shape," which are estimated to have a negative impact on the evaluation of the outfit or fashion item, have not been taken into consideration. To give a specific example, the information processing device 10 provides content relating to outfits or fashion items in which the attached elements "head" and / or "body shape," which are not taken into consideration, using a model that has been trained as training data, based on rankings determined based on evaluation results relating to target information in which the attached elements "head" and / or "body shape," which are estimated to have a negative impact on the evaluation of the evaluation target, which is the outfit or fashion item. To give a more specific example, when the information processing device 10 receives input consisting of a first target group, which is a combination of some of the fashion items that make up the outfit to be evaluated; a second target group, which is a combination of the remaining parts of the fashion items that make up the outfit to be evaluated; and an additional target other than the "head" and / or "body shape" that is estimated to have a negative impact on the evaluation of the outfit to be evaluated (for example, the additional target "pose"), the information processing device 10 calculates a set matching score using a model that has been trained to output a higher set matching score the higher the degree of matching between the first target group, the second target group, and the additional target (the closer it is to a predetermined combination of a predetermined evaluation target (which consists of a predetermined first target group and a predetermined second target group) and a predetermined additional target that has a higher ranking. The information processing device 10 then provides content related to the combination of the first target group, the second target group, and the additional information that is to be calculated, based on the calculated set matching score.For example, the information processing device 10 may provide content relating to outfits that are combinations of a first target group and a second target group that are subject to calculation, taking into account the additional item "pose" that is subject to calculation, or content relating to multiple fashion items that constitute the outfit, which is a combination of a first target group and a second target group that are subject to calculation, taking into account the additional item "pose" that is subject to calculation. In addition, the information processing device 10 may provide content relating to the additional item "pose" that is subject to calculation and matches with the outfit, which is a combination of a first target group and a second target group that are subject to calculation, or content relating to the degree of matching (set matching score) between the outfit, which is a combination of a first target group and a second target group that are subject to calculation, and the additional information, the additional item "pose" that is subject to calculation.

[0071] Furthermore, in evaluating an outfit or fashion item, the information processing device 10 may prioritize providing content in which an attachment presumed to have a positive influence is attached to the outfit or fashion item, over content in which an attachment presumed to have a negative influence is attached to the outfit or fashion item.

[0072] As described above, the information processing device 10 according to the embodiment can understand the influence on the evaluation of an outfit, not only on the compatibility of the clothing that makes up the outfit, but also on the influence on the evaluation of additional elements such as head shape and body shape. Similarly, the information processing device 10 according to the embodiment can understand the influence on the evaluation of a fashion item, such as the influence on the evaluation of a fashion item, such as the influence on the evaluation of a fashion item. In other words, the information processing device 10 according to the embodiment can understand the influence on the evaluation of an item that has been added to it.

[0073] Furthermore, the information processing device 10 according to the embodiment can prioritize providing the user with content to which additional elements are attached that are presumed to have a positive influence on the evaluation of the outfit or fashion item, and if such additional elements are not attached, it can provide content to which such additional elements are attached. In other words, the information processing device 10 according to the embodiment can provide content related to the evaluation target in accordance with the influence that the elements attached to the evaluation target have on the evaluation of the evaluation target.

[0074] Furthermore, in the above embodiment, if there are evaluators who are not serious about evaluating the target information, the reliability of ratings and the like will decrease, and there is a possibility that inappropriate target information will be selected when selecting the target information to present to the evaluator. Therefore, the information processing device 10 according to the embodiment can set appropriate ratings based on evaluation results from other serious evaluators, in order to exclude evaluation results from evaluators who are presumed to be not serious, and can appropriately select the target information to present to the evaluator. In other words, the information processing device 10 according to the embodiment can select the information to present to the evaluator according to the content of the evaluator's evaluation.

[0075] [3. Other processing examples] The process described above is merely one example, and the information processing device 10 may perform various processes using various types of information. Examples of this are listed below.

[0076] [3-1. Regarding the subjects of the additional requirements] In the example shown in Figure 2, the information processing device 10 may select and present to the evaluator target information to which additional elements have been added, such as the pose of the person wearing the outfit, the way the clothes constituting the outfit are worn (in other words, the manner in which they are worn), a background image, a description of the scene in which the outfit is worn, a description of the weather in which it is worn, and outfits worn by companions of the person wearing the outfit. For example, the information processing device 10 selects an image of a person wearing the outfit and striking a predetermined pose, showing only the part of the person from the head down (an image that does not include the head; the body shape may also be excluded), as target information to which the additional element "pose" has been added, and presents it to the evaluator. Alternatively, the information processing device 10 selects an image of a person wearing the outfit and in a predetermined manner, showing only the part of the person from the head down (an image that does not include the head; the body shape may also be excluded), as target information to which the additional element "dressing style" has been added, and presents it to the evaluator. Furthermore, the information processing device 10 selects an image in which an image of the outfit is superimposed on a predetermined background image (which may be an image that excludes the head and body shape of the person wearing the outfit) as target information to which the attachment target "background image" is attached, and presents it to the evaluator. Furthermore, the information processing device 10 selects content that explains the occasion and weather when the outfit is worn, along with an image of the person wearing the outfit (which may be an image that excludes the head and body shape), as target information to which the attachment targets "occasion" and "weather" are attached, and presents it to the evaluator. Furthermore, the information processing device 10 selects an image of a companion (which may be an image that excludes the head and body shape) along with an image of the person wearing the outfit (which may be an image that excludes the head and body shape), as target information to which the attachment target "companion" is attached, and presents it to the evaluator. Finally, the information processing device 10 provides content related to the outfit to user U1 based on the influence that the attachment targets "pose," "style," "background image," "occasion," "weather," and "companion" have on the evaluation of the outfit.

[0077] Furthermore, an image of a person wearing the outfit and standing perfectly still may be considered as target information without the "pose" attribute attached, while an image in which some change has been made from that posture (for example, bending the arms or pulling the legs back) may be considered as target information with the "pose" attribute attached.

[0078] Furthermore, images of a person wearing the outfit may be treated as target information without the attached "style" attribute, while images in which some variation has been added to the way the outfit is worn (for example, the person is wearing the outfit as a jacket rather than a dress, the buttons of the outfit are not fastened, or the outfit is tied around the waist) may be treated as target information with the attached "style" attribute.

[0079] Furthermore, an image showing an outfit with a plain background (for example, a white background) may be considered as target information without the attached "background image," while an image in which some kind of change has been made to the background (for example, changing from a white background to an image of a party venue, or changing from a white background to an image of the sea) may be considered as target information with the attached "background image."

[0080] Furthermore, the information processing device 10 may evaluate a coordinate or fashion item represented in grayscale, and estimate whether the color of the fashion item has a positive or negative influence on the evaluation of the coordinate or fashion item.

[0081] Furthermore, when "head" or "body type" is used as an attachment target, the information processing device 10 may either fix the person and prepare attachment information with the same "head" or "body type" attached, or it may prepare attachment information with different "heads" or "body types" attached without fixing the person. If attachment information with different "heads" or "body types" is prepared without fixing the person, the information processing device 10 may treat the different "heads" or "body types" as attachment information with the same type of attachment target "head" or "body type," or it may treat them as attachment information with different types of attachment targets "head A" or "body type A" and "head B" or "body type B." If attachment information with different types of attachment targets "head A" or "body type A" and "head B" or "body type B" is treated, the information processing device 10 may determine the respective rankings based on the evaluation results for each piece of attachment information. The information processing device 10 may then calculate a score indicating the degree of agreement between each ranking and estimate the influence of the attachment target.

[0082] [3-2. Estimation of the degree of impact] In the example shown in Figure 2, the information processing device 10 may estimate that the greater the difference between score Sc1 and score Sc2, the greater the influence that the added item "head" has on the evaluation of the outfit. To give a specific example, the information processing device 10 estimates that the greater the positive influence that the added item "head" has on the evaluation of the outfit, the higher the score Sc2 is than score Sc1, and the greater the negative influence that the added item "head" has on the evaluation of the outfit, the lower the score Sc2 is than score Sc1.

[0083] [3-3. About the outfit] In the example shown in Figure 2, the information processing device 10 may select target information to present to the evaluator from among the target information that represents a code identified using Model #2, which is pre-generated using posted information on a predetermined web service (e.g., a coordination service) as training data (ground truth data). The information processing device 10 may then train Model #2 based on the evaluation of the target information. For example, the information processing device 10 may train Model #2 using codes whose rating is above a predetermined threshold as ground truth data.

[0084] Furthermore, the information processing device 10 may train model #2 depending on whether or not the added item has an effect on the evaluation of the outfit. For example, if the added item "head" has an effect on the evaluation of the outfit, the information processing device 10 will train model #2 with images of outfits to which the added item "head" has been added. If the added item "head" does not have an effect on the evaluation of the outfit, the information processing device 10 will train model #2 with images of outfits to which the added item "head" has not been added.

[0085] This means that, for example, if the attached "head" affects the evaluation of the outfit (in other words, if the attached "head" is necessary for evaluating the outfit), and the attached "head" is not attached to the ground truth data, the accuracy of model #2 can be improved by attaching the attached "head" and training model #2. Also, if the attached "head" does not affect the evaluation of the outfit (in other words, if the attached "head" is not necessary for evaluating the outfit), images of outfits without the attached "head" can be used as ground truth data, thus reducing the effort required to prepare ground truth data and improving convenience.

[0086] [3-4. Regarding the subdivision of the target of the addition] In the example shown in Figure 2, the information processing device 10 may select and present to the evaluator some of the additional elements that are estimated to have an impact on the evaluation of the outfit, receive an evaluation of the said information, and estimate the impact of those additional elements on the evaluation of the outfit. For example, if the additional element "body shape" is estimated to have an impact on the evaluation of the outfit, the information processing device 10 selects and presents to the evaluator the information with the additional element "body shape: excluding hands" added. To give a specific example, the information processing device 10 presents to the evaluator an image of a person wearing the outfit from the head down, excluding the hands (the hands have been removed), as the information with the additional element "body shape: excluding hands" added. Then, based on the evaluation from the evaluator, the information processing device 10 estimates the impact of the additional element "body shape: excluding hands" on the evaluation of the outfit.

[0087] In this way, by subdividing the elements that influence the evaluation of an outfit and estimating their influence, it is possible to accurately understand which parts of the elements have an impact.

[0088] Furthermore, the attached items may include, for example, "Body type: Hands only" which shows only the hands, "Body type: Feet only" which shows only the feet, or "Body type: Waist only" which shows only the waist area. Also, the attached items may include "Head: Non-hair" which shows everything except the hair, or "Head: Hair only" which shows only the hair.

[0089] [3-5. About the Content] In the example shown in Figure 2, the information processing device 10 may provide content to which an additional element is added that is estimated to have a positive influence on the user U1's evaluation of the outfit, provided that the additional element satisfies predetermined conditions. For example, if the additional element "background image" is estimated to have a positive influence on the evaluation of the outfit, and it is estimated that user U1 is considering purchasing clothing for a party based on a questionnaire or the like, the information processing device 10 provides content to which a party image is added as the background image of the outfit.

[0090] Furthermore, the information processing device 10 may provide content based on the user U1's purchase history. For example, if the added "head" is estimated to have a positive influence on the evaluation of the outfit, the information processing device 10 will provide content in which an image of a head corresponding to the product image of the clothing purchased by user U1 in an e-commerce service, etc., is added to the outfit. In other words, the information processing device 10 will provide content in which an image of a head that is estimated to be preferred by user U1 is added to the outfit. Alternatively, user U1 may register their head image, and the information processing device 10 may provide content in which user U1's head image is added to the product image of the clothing purchased by user U1 in an e-commerce service, etc.

[0091] Furthermore, the information processing device 10 may provide content based on the user U1's browsing history. For example, if it is estimated that the added element "face" has a positive influence on the evaluation of the outfit, the information processing device 10 will provide content in which the image of the outfit that user U1 has reacted positively to (for example, by clicking the "Like" button) in the coordination service, etc., and the corresponding head image (for example, the head image of the person who reacted positively) are added to the outfit. In other words, the information processing device 10 will provide content in which the head image that user U1 is estimated to like is added to the outfit. Alternatively, user U1 may register their head image, and the information processing device 10 may provide content in which user U1's head image is added to the image of the outfit that user U1 has reacted positively to in the coordination service, etc.

[0092] [3-6. Regarding the exclusion of evaluation results] In the example shown in Figure 2, the information processing device 10 may present dummy information along with the target information, and if the evaluator rates the dummy information favorably, the evaluator's evaluation result may be excluded. For example, the information processing device 10 may present dummy information, such as an image containing the text "Do not select this image," along with the target information, and if the evaluator rates the dummy information favorably, the evaluator's evaluation result may be excluded.

[0093] Furthermore, if the information processing device 10 presents two identical pieces of information and the evaluator rates either one favorably, the evaluator's evaluation result may be excluded.

[0094] Furthermore, the information processing device 10 may present correct target information (for example, target information that is clearly a valid outfit to anyone) and incorrect target information (for example, target information that is clearly not a valid outfit to anyone), and if an evaluator rates the incorrect target information favorably, the evaluator's rating may be excluded.

[0095] Furthermore, the information processing device 10 may exclude the evaluation result made by the evaluator if the time from when the target information is displayed until the evaluator makes a selection is greater than or equal to a predetermined threshold.

[0096] Furthermore, if the information processing device 10 presents the target information to the evaluator via an e-commerce service or a coordination service, and while the information processing device 10 is presenting the information, an application related to another service (e.g., a messaging app) or a webpage is displayed on the evaluator's terminal 200, the evaluation result of that evaluator may be excluded. However, even if an application related to another service (e.g., a messaging app) or a webpage is displayed on the evaluator's terminal 200, the information processing device 10 does not need to exclude the evaluation result of that evaluator if the time until the evaluator returns to the evaluation is below a predetermined threshold, or if the time from when the evaluator returns to the evaluation until they make a selection is below a predetermined threshold.

[0097] Furthermore, the information processing device 10 may exclude the evaluation result of an evaluator if the time from when the target information is displayed until the evaluator makes a selection is below a predetermined threshold. However, instead of immediately excluding the result, the information processing device 10 may encourage the evaluator to evaluate seriously. For example, if the time from when the target information is displayed until the evaluator makes a selection is below a predetermined threshold, the information processing device 10 may control the time from when the target information is displayed until the evaluator can make a selection to be longer than usual. This ensures that the evaluator has enough time to carefully review the displayed target information, and also controls the system so that those who evaluate seriously can finish faster, thus encouraging the evaluator to answer seriously.

[0098] Furthermore, the information processing device 10 may exclude an evaluator's evaluation results if the evaluator's evaluation content meets conditions set based on information about the evaluator. For example, if, based on information about the evaluator such as a questionnaire to the evaluator, gender, age, place of residence, clothing purchase history, and outfit browsing history, it is estimated that the fashion category the evaluator is not good at (in other words, the fashion category the evaluator does not usually wear) is "mode," then it is estimated that evaluating target information that represents an outfit belonging to "mode" will take time. Therefore, in evaluating target information that represents an outfit belonging to "mode," the information processing device 10 will not exclude the evaluator's evaluation results even if the time from when the target information is displayed until the evaluator selects it exceeds a predetermined threshold. On the other hand, in evaluating target information that represents an outfit belonging to "mode," the information processing device 10 will exclude the evaluator's evaluation results if the time from when the target information is displayed until the evaluator selects it exceeds a specific threshold that is longer than a predetermined threshold.

[0099] Furthermore, if the evaluator's preferred fashion category (in other words, the fashion category the evaluator usually wears) is estimated to be "casual," it is estimated that evaluating target information representing outfits belonging to "casual" will not take much time. Therefore, when evaluating target information representing outfits belonging to "casual," the information processing device 10 sets a shorter time limit from when the target information is displayed until the evaluator makes a selection than that for evaluators who are not proficient in "casual," and if the time limit is exceeded, the evaluation result by that evaluator is excluded.

[0100] Furthermore, if the information processing device 10 sets an evaluator who evaluates outfits #1 to #25 as an excluded person, it may exclude the evaluation results that the excluded person has previously made for other outfits. For example, if the excluded person has previously evaluated outfits #26 to #50, the information processing device 10 will exclude the evaluation results of the excluded person, determine the ranking of outfits #26 to #50 based on the evaluation results of other evaluators, and estimate the influence that the added elements attached to outfits #26 to #50 have on the evaluation.

[0101] [3-7. Regarding excluded persons] In the example in Figure 2, the information processing device 10 may decide whether or not to present target information to the excluded person (in other words, whether or not to reinstate them as an evaluator) based on information about the excluded person. For example, the excluded person may have given an unserious evaluation last time because they were presented with target information about a fashion category they were not good at (in other words, they may have given a serious evaluation if they had been presented with target information about a fashion category they were good at). Therefore, the information processing device 10 decides whether or not to reinstate the excluded person as an evaluator based on information about the excluded person. To give a specific example, if target information about a fashion category that the excluded person is presumed to be bad at is presented based on information about the excluded person, and the excluded person was set as an excluded person based on their evaluation of that target information, the information processing device 10 will reinstate the excluded person as an evaluator. In addition, the information processing device 10 decides whether or not to reinstate the excluded person as an evaluator based on the content of the excluded person's evaluation of other target information. To give a specific example, if target information about a fashion category that the excluded person is presumed to be good at is presented based on information about the excluded person, and the excluded person was not set as an excluded person based on their evaluation of that target information (in other words, if it is determined that they gave a serious evaluation), the excluded person will be reinstated as an evaluator. Furthermore, the information processing device 10 may reinstate the excluded person to the evaluator role only when it presents relevant information relating to a fashion category in which the excluded person is presumed to be an expert.

[0102] Furthermore, the information processing device 10 may present predetermined information and, based on the evaluation results for said predetermined information, decide whether or not to reinstate the excluded person as an evaluator. For example, the information processing device 10 may present the excluded person with both correct and incorrect target information a predetermined number of times, and if the percentage of the person who selected the correct target information is above a predetermined threshold, the excluded person may be reinstated as an evaluator.

[0103] Furthermore, when the information processing device 10 evaluates target information related to a fashion category in which the excluded person is an expert, it may reinstate the excluded person as an evaluator.

[0104] Furthermore, if the information processing device 10 presents the target information to the evaluator corresponding to the attributes possessed by the excluded person, it may reinstate the excluded person to the role of evaluator.

[0105] [3-8. Regarding Evaluation] In the example shown in Figure 2, the information processing device 10 presents the evaluator with two pieces of information and accepts an evaluation indicating which piece of information is more favorable. However, the evaluator's evaluation is not limited to this example and may be conducted in any manner. For example, when the information processing device 10 presents pieces of information #1 and #2, it may present options such as (1) "Piece of information #1 is more favorable", (2) "Piece of information #1 is somewhat more favorable", (3) "Piece of information #2 is more favorable", (4) "Piece of information #2 is somewhat more favorable", or (5) "Both are equally favorable", and accept a selection from the evaluator.

[0106] Furthermore, if (2) is selected, the information processing device 10 may set the value used to increase the rating of target information #1 to be lower than if (1) were selected. Also, if (4) is selected, the value used to increase the rating of target information #2 may be lower than if (3) were selected. In addition, if (5) is selected, the information processing device 10 does not need to change the ratings of target information #1 and #2.

[0107] Furthermore, instead of accepting an evaluation indicating which of the two pieces of information is favorable, the information processing device 10 may present three or more pieces of information and accept an evaluation indicating which one is favorable.

[0108] [3-9. About the Group] In the example in Figure 2, the classification of evaluators into groups may be arbitrary. For example, the information processing device 10 may not classify evaluators into groups, but instead select and present the target information to each evaluator based on the overall rating and number of evaluations of the target information among all evaluators, and receive evaluations of the target information from each evaluator. The information processing device 10 may then extract evaluators who have attributes corresponding to user U1, who is the target of the content, and estimate the impact of the added items on the evaluation of the outfit (in other words, the impact of the added items on user U1's evaluation of the outfit) based on the evaluation results from the extracted evaluators. The information processing device 10 may then provide user U1 with content related to the outfit based on the impact of the added items on the outfit.

[0109] Furthermore, the information processing device 10 may select and present the target information to each evaluator based on the overall rating and number of evaluations of the target information among all evaluators, without classifying the evaluators into groups, and may receive evaluations of the target information from each evaluator. The information processing device 10 may also extract evaluators who have attributes specified by the administrator of an e-commerce service or a coordination service, and estimate the impact of the added item on the evaluation of the coordination (in other words, the impact of the added item on the evaluation of the coordination in the specified attributes) based on the evaluation results from the extracted evaluators. The information processing device 10 may also provide the administrator with content regarding the impact of the added item on the coordination.

[0110] In the example in Figure 2, the classification of evaluators into groups may be performed dynamically. For example, the information processing device 10 may not classify evaluators into groups, but instead select and present the target information to each evaluator based on the overall rating and number of evaluations of the target information among all evaluators, and receive evaluations of the target information from each evaluator. Alternatively, for example, the information processing device 10 may randomly classify evaluators into groups, and then select and present the target information to each evaluator based on the rating and number of evaluations of the target information in each group, and receive evaluations of the target information from each evaluator. Furthermore, the information processing device 10 may, at some point (when a predetermined number of evaluations have been received, for example, when each evaluator's evaluation trend has emerged), move the groups based on the similarity of each evaluator's evaluation trend (also called each evaluator's attribute) so that evaluators with the same or similar evaluation trends belong to the same group (at this time, new groups may be generated according to the number of evaluation trends that have emerged), and then select and present the target information to each evaluator based on the rating and number of evaluations of the target information in each group after the move, and receive evaluations of the target information from each evaluator. Here, evaluators with similar evaluation tendencies are, for example, evaluators whose evaluations of the same outfit are similar, or whose evaluations of similar outfits (for example, outfits that include the same clothing items, outfits with similar appearance such as color and shape, outfits in the same category, outfits with similar prices, etc.) are similar.

[0111] The information processing device 10 may identify a group of evaluators who have attributes corresponding to the user U1 to whom the content is provided, determine a ranking based on the evaluation results in the identified group, and estimate the influence of the added item on the evaluation of the outfit (in other words, the influence of the added item on the evaluation of the outfit by user U1). The information processing device 10 may then provide user U1 with content related to the outfit based on the influence of the added item on the outfit.

[0112] In the example in Figure 2, the classification of evaluators into groups may be performed dynamically. For example, the information processing device 10 may not classify evaluators into groups, but instead select and present the target information to each evaluator based on the overall rating and number of evaluations of the target information among all evaluators, and receive evaluations of the target information from each evaluator. Alternatively, for example, the information processing device 10 may randomly classify evaluators into groups, and then select and present the target information to each evaluator based on the rating and number of evaluations of the target information within each group, and receive evaluations of the target information from each evaluator. Furthermore, the information processing device 10 may, at some point, select and present target information suitable for estimating the similarity of the evaluators' evaluation tendencies (also called attributes), and receive evaluations of the target information from each evaluator. Specifically, the information processing device 10 may select and present a pair of target information such that, if the evaluator's evaluation tendency is a predetermined evaluation tendency, the left-hand target information will definitely be rated as preferable. It may then move the groups so that evaluators with the same or similar evaluation tendencies belong to the same group (at this time, new groups may be generated depending on the number of evaluation tendencies that appear). Based on the ratings and evaluation counts of the target information in each group after the move, it may select and present the target information to each evaluator and receive evaluations of the target information from each evaluator. The information processing device 10 may then identify the groups to which evaluators with attributes corresponding to the user U1, who is the target of the content, belong. Based on the evaluation results in the identified groups, it may determine a ranking and estimate the impact of the added item on the evaluation of the outfit (in other words, the impact of the added item on the evaluation of the outfit by user U1). The information processing device 10 may then provide content related to the outfit to user U1 based on the impact of the added item on the outfit.

[0113] [3-10. About the evaluators] The information processing device 10 may, on behalf of the evaluator, present multiple pieces of target information to an artificial intelligence such as a large-scale language model and accept its evaluation of the target information. In principle, the information processing device 10 presents multiple pieces of target information to the artificial intelligence on behalf of the evaluator and accepts its evaluation of the target information. However, if it cannot receive a correct evaluation from the artificial intelligence, it may present the same multiple pieces of target information to the evaluator and accept its evaluation of the target information. Furthermore, the information processing device 10 may present the multiple pieces of target information to the artificial intelligence in a first order and accept its evaluation, and then present the same multiple pieces of target information to the artificial intelligence in a second order and accept its evaluation. If the two evaluations are the same, it may be judged as a correct evaluation; if the two evaluations are different, it may be judged as an incorrect evaluation (an evaluation influenced in some way by the order). Furthermore, the information processing device 10 presents multiple pieces of information to the artificial intelligence, accepts evaluations of the information and reasons for the evaluation, and may judge the evaluation as correct if the reason for the evaluation is valid (for example, because the outfit presented on the left appears to have more harmonious clothing), or it may judge the evaluation as incorrect if the reason for the evaluation is not valid (for example, simply because it is presented on the left). In addition, when having the artificial intelligence make an evaluation, the information processing device 10 may determine and instruct the artificial intelligence to have attributes so that the evaluator has attributes (for example, it may generate a prompt instructing it to evaluate as a male in his 20s), or it may allow the artificial intelligence to determine its own attributes (for example, it may generate a prompt instructing it to evaluate from a position determined by the artificial intelligence itself).

[0114] [4. Configuration of Information Processing Device] Next, the configuration of the information processing device 10 will be described using Figure 5. Figure 5 is a diagram showing an example of the configuration of the information processing device 10 according to the embodiment. As shown in Figure 5, the information processing device 10 has a communication unit 20, a storage unit 30, and a control unit 40.

[0115] (Regarding Communications Section 20) The communication unit 20 is implemented, for example, by a NIC (Network Interface Card). The communication unit 20 is connected to the network N by wire or wireless connection and transmits and receives information between it and the user terminal 100, the evaluator terminal 200, etc.

[0116] (Regarding memory unit 30) The storage unit 30 is implemented by, for example, semiconductor memory elements such as RAM (Random Access Memory) and flash memory, or by storage devices such as hard disks and optical discs. As shown in Figure 5, the storage unit 30 has a target information database 31, an evaluator information database 32, and a user information database 33.

[0117] (Regarding the target information database 31) The target information database 31 stores various types of information related to the evaluation target. Here, an example of the information stored in the target information database 31 will be explained using Figure 6. Figure 6 is a diagram showing an example of the target information database 31 according to the embodiment. In the example in Figure 6, the target information database 31 has items such as "Target Information ID", "Evaluation Target ID", "Evaluation Target Information", "Additional Target", "Rating", and "Evaluation Count".

[0118] "Target Information ID" indicates identification information for identifying the target information. "Evaluation Target ID" indicates identification information for identifying the evaluation target indicated by the target information. "Evaluation Target Information" indicates information about the evaluation target indicated by the target information, for example, an image of the evaluation target is stored. "Additional Item" indicates additional items attached to the evaluation target, for example, an image of the head of the person wearing the evaluation target, information (including images) about the body shape of the person wearing the evaluation target, the pose of the person wearing the evaluation target, the manner in which the evaluation target is worn, and background images are stored. "Rating" indicates the rating of the target information. "Evaluation Count" indicates the number of times the target information has been evaluated.

[0119] In other words, Figure 6 shows an example where the target information identified by the target information ID "DID#1" is identified by the target evaluation ID "CID#1", the target evaluation information for that evaluation is "Target Evaluation Information #1", the added item is "Added Item #1", the rating of that target information is "Rating #1", and the number of evaluations is "Number of Evaluations #1".

[0120] (Regarding the evaluator information database 32) The evaluator information database 32 stores various types of information about evaluators. Here, an example of the information stored in the evaluator information database 32 will be explained using Figure 7. Figure 7 is a diagram showing an example of the evaluator information database 32 according to the embodiment. In the example in Figure 7, the evaluator information database 32 has items such as "evaluator ID," "attribute information," "purchase history," "browsing history," and "evaluation information."

[0121] "Evaluator ID" indicates identification information used to identify the evaluator. "Attribute Information" indicates the evaluator's attributes. "Purchase History" shows the evaluator's purchase history in e-commerce services, etc. "Browsing History" shows the evaluator's content browsing history in e-commerce services, coordination services, etc.

[0122] "Evaluation information" refers to information about the evaluator's evaluation of the target information, and includes items such as "Presented Information" and "Evaluation." "Presented Information" refers to information about the target information that was presented to the evaluator (in other words, the information that the evaluator evaluated). "Evaluation" refers to the evaluator's evaluation of the target information presented to them, and for example, it stores information indicating which of the target information was viewed favorably.

[0123] In other words, Figure 7 shows an example where the attribute information of the evaluator identified by the evaluator ID "AID#1" is "Attribute Information #11", the purchase history is "Purchase History #11", the browsing history is "Browsing History #11", the target information presented to the evaluator is "Presented Target Information #1", and the evaluation of that target information is "Evaluation #1".

[0124] (Regarding User Information Database 33) The user information database 33 stores various types of information about the user. Here, an example of the information stored in the user information database 33 will be explained using Figure 8. Figure 8 is a diagram showing an example of the user information database 33 according to the embodiment. In the example in Figure 8, the user information database 33 has items such as "User ID," "Attribute Information," "Purchase History," and "Browsing History."

[0125] "User ID" indicates identification information used to identify the user. "Attribute Information" indicates the user's attributes. "Purchase History" shows the user's purchase history in e-commerce services, etc. "Browsing History" shows the user's content browsing history in e-commerce services, coordination services, etc.

[0126] In other words, Figure 8 shows an example where the attribute information of a user identified by the user ID "UID#1" is "Attribute Information#21", the purchase history is "Purchase History#21", and the browsing history is "Browsing History#21".

[0127] (Regarding the control unit 40) The control unit 40 is a controller, and is realized, for example, by a CPU (Central Processing Unit) or MPU (Micro Processing Unit) executing various programs stored in the memory device inside the information processing device 10 using RAM as a working area. Alternatively, the control unit 40 is a controller, and is realized, for example, by an integrated circuit such as an ASIC (Application Specific Integrated Circuit) or FPGA (Field Programmable Gate Array). As shown in Figure 5, the control unit 40 according to this embodiment has a selection unit 41, a reception unit 42, an estimation unit 43, a provision unit 44, a setting unit 45, a determination unit 46, and a learning unit 47, and realizes or executes the information processing functions and operations described below.

[0128] (Regarding selection section 41) The selection unit 41 selects multiple first target information and multiple second target information to present to the evaluator from among first target information that shows only the items to be evaluated and second target information that shows additional items along with the items to be evaluated. For example, in the example in Figure 2, the selection unit 41 refers to the target information database 31 and the evaluator information database 32 and selects target information to present to the evaluator from target information group #1 (first target information), which consists of images showing only outfits #1 to #25. The selection unit 41 also selects target information to present to the evaluator from target information group #2 (second target information), which shows images of the heads of people in addition to the images of outfits #1 to #25. Furthermore, the selection unit 41 selects two pieces of target information to present to the evaluator from target information group #3 (second target information), which consists of images showing the body shape of people wearing outfits #1 to #25. Furthermore, the selection unit 41 selects target information to present to the evaluator from the target information group #4 (second target information), which consists of images showing people wearing outfits #1 to #25.

[0129] Furthermore, the selection unit 41 may select multiple first target information and multiple second target information to present to the evaluator from among first target information, which is an image showing only combinations of clothing, and second target information, which is an image showing additional objects along with combinations of clothing. For example, in the example in Figure 2, the selection unit 41 selects target information to present to the evaluator from target information group #1, which is an image showing only combinations of clothing, such as outfits #1 to #25. The selection unit 41 also selects target information to present to the evaluator from target information group #2, which shows images of a person's head in addition to the images of combinations of clothing, such as outfits #1 to #25. The selection unit 41 also selects target information to present to the evaluator from target information group #4, which is an image showing a person wearing combinations of clothing, such as outfits #1 to #25.

[0130] Furthermore, the selection unit 41 may select multiple second target information items to present to the evaluator from among the second target information items that indicate at least one of the following as additional targets: an image of a person's head, information about the body shape of the person using the item to be evaluated, the pose of the person using the item to be evaluated, the manner in which the item to be evaluated is used, the situation in which the item to be evaluated is used (more specifically, TPO (Time Place Occasion)), and a background image. For example, in the example in Figure 2, the selection unit 41 selects target information to present to the evaluator from target information group #2, which further indicates an image of a person's head in addition to the images of outfits #1 to #25. The selection unit 41 also selects two pieces of target information to present to the evaluator from target information group #3, which consists of images showing the body shape of a person wearing outfits #1 to #25. The selection unit 41 also selects target information to present to the evaluator from target information group #4, which consists of images showing a person wearing outfits #1 to #25.

[0131] Furthermore, the selection unit 41 may select multiple first target information items to present to the evaluator from among the first target information items based on the rating of the first target information items calculated based on the first evaluation, and select multiple second target information items to present to the evaluator from among the second target information items based on the rating of the second target information items calculated based on the second evaluation. For example, in the example in Figure 2, the selection unit 41 selects target information to present to the evaluator based on the rating of each target information item included in target information group #1. Similarly, the selection unit 41 selects target information to present to the evaluator from target information groups #2 to #4.

[0132] Furthermore, the selection unit 41 may select multiple first target information items from the first target information items whose rating differences are within a predetermined range, and select multiple second target information items from the second target information items whose rating differences are within a predetermined range. For example, in the example in Figure 2, the selection unit 41 selects target information items whose rating difference with target information item #1 is within a predetermined range, based on the rating of each target information item included in target information group #1. Similarly, the selection unit 41 selects target information items to present to the evaluator from target information groups #2 to #4.

[0133] Furthermore, the selection unit 41 may select multiple first target information items to present to the evaluators from among the first target information items based on the number of times the first target information items have been evaluated by the evaluators, and may select multiple second target information items to present to the evaluators from among the second target information items based on the number of times the second target information items have been evaluated by the evaluators. For example, in the example in Figure 2, the selection unit 41 selects target information item #1 from target information group #1 that currently has the fewest evaluation counts, indicating that it has been evaluated by each evaluator belonging to group G1. Similarly, the selection unit 41 selects target information items to present to the evaluators from target information groups #2 to #4.

[0134] Furthermore, the selection unit 41 may select multiple first target information and multiple third target information to present to the evaluator from among first target information that shows only the evaluation target, and third target information that shows some of the additional targets that are estimated to have an impact on the evaluation of the evaluation target, together with the evaluation target. For example, in the example in Figure 2, if the additional target "body shape" is estimated to have an impact on the evaluation of the outfit, the selection unit 41 selects the target information to present to the evaluator from among the target information (third target information) to which the additional target "body shape: excluding hands" has been added.

[0135] Furthermore, the selection unit 41 may select multiple first target information items to present to the evaluator from among the first target information items that indicate combinations of items to be evaluated, which have been identified using a model trained to determine whether or not a combination of items to be evaluated satisfies predetermined conditions, and may also select multiple second target information items to present to the evaluator from among the second target information items that indicate combinations of items to be evaluated, which have been identified using the same model. For example, in the example in Figure 2, the selection unit 41 selects target information items to present to the evaluator from among the target information items that indicate codes identified using model #2.

[0136] Furthermore, the selection unit 41 may select multiple pieces of target information to present to the evaluators from among the target information indicating the items to be evaluated. The selection unit 41 may also select new target information based on the evaluation results obtained by evaluators, excluding the evaluation results of evaluators whose evaluation content for the target information meets predetermined conditions. For example, in the example in Figure 2, the selection unit 41 excludes the evaluation results of excluded evaluators and selects target information indicating the items to be evaluated to present to evaluators who are not designated as excluded evaluators.

[0137] Furthermore, if the evaluator evaluates the dummy information presented with the target information, or the dummy information presented in place of the target information, as favorable, the selection unit 41 may select new target information based on the evaluation results excluding the evaluator's evaluation. For example, in the example in Figure 2, if the selection unit 41 presents dummy information with the target information and the evaluator evaluates the dummy information as favorable, the selection unit 41 excludes the evaluator's evaluation and selects new target information. In addition, if the selection unit 41 presents dummy information in place of the target information and the evaluator evaluates the dummy information as favorable, the selection unit 41 may exclude the evaluator's evaluation and select new target information.

[0138] Furthermore, if the evaluator rates either of the same pieces of target information favorably, the selection unit 41 may select new target information based on the evaluation results excluding the evaluator's evaluation. For example, in the example in Figure 2, if the selection unit 41 presents two pieces of the same target information and the evaluator rates either of them favorably, it excludes the evaluator's evaluation and selects new target information.

[0139] Furthermore, if the evaluator evaluates a pre-set target information as unfavorable, the selection unit 41 may select new target information based on the evaluation results excluding the evaluator's evaluation. For example, in the example in Figure 2, the selection unit 41 presents target information that is correct and target information that is incorrect. If the evaluator evaluates the incorrect target information favorably, the selection unit 41 excludes the evaluator's evaluation and selects new target information.

[0140] Furthermore, the selection unit 41 may select new target information based on evaluation results that exclude the evaluation results of the evaluator, provided that the time taken by the evaluator to evaluate the target information satisfies predetermined conditions. For example, in the example in Figure 2, if the time from when the target information is displayed until the evaluator makes a selection is greater than or equal to a first threshold, the selection unit 41 excludes the evaluation results of the evaluator and selects new target information. Also, if the time from when the target information is displayed until the evaluator makes a selection is less than or equal to a second threshold, the selection unit 41 excludes the evaluation results of the evaluator and selects new target information.

[0141] Furthermore, the selection unit 41 may select new target information based on evaluation results that exclude the evaluation results made by the evaluator, if the location on the screen of the terminal device used by the evaluator where the evaluator performed an operation related to the evaluation of the target information satisfies predetermined conditions. For example, in the example of Figure 2, the selection unit 41 excludes the evaluation results made by the evaluator and selects new target information if the evaluator has always selected target information displayed in a predetermined area on the screen of the evaluator terminal 200, or if the evaluator has selected target information by pressing an area on the screen of the evaluator terminal 200 in a certain pattern.

[0142] Furthermore, if the evaluator performs an operation on the terminal device used by the evaluator to display a screen different from the screen used to evaluate the target information, the selection unit 41 may select new target information based on evaluation results that exclude the evaluator's evaluation results. For example, in the example in Figure 2, if the selection unit 41 presents target information to the evaluator via an e-commerce service or a coordination service, and while the information is being presented, an app or web page related to another service is displayed on the evaluator's terminal 200, the selection unit 41 excludes the evaluator's evaluation results and selects new target information.

[0143] Furthermore, the selection unit 41 may select new target information based on evaluation results that exclude the evaluation result of the evaluator, if the evaluator's evaluation of the target information satisfies conditions set based on information about the evaluator. For example, in the example in Figure 2, if it is estimated that the evaluator's weakest fashion category is "mode," the selection unit 41 will not exclude the evaluator's evaluation result in the evaluation of target information representing outfits belonging to "mode," even if the time from when the target information is displayed until the evaluator makes a selection exceeds a predetermined threshold. Also, if it is estimated that the evaluator's strongest fashion category is "casual," the selection unit 41 will set a shorter time limit from when the target information is displayed until the evaluator makes a selection in the evaluation of target information representing outfits belonging to "casual," than that of evaluators who are not strong in "casual," and if the time limit is exceeded, it will exclude the evaluator's evaluation result and select new target information.

[0144] Furthermore, the selection unit 41 may select new target information based on evaluation results excluding the evaluator's evaluation result if the evaluator's evaluation of the target information satisfies conditions set based on the evaluator's purchase history of the items being evaluated. For example, in the example in Figure 2, the selection unit 41 selects new target information based on evaluation results excluding the evaluator's evaluation result if the evaluator's evaluation of the target information satisfies conditions set based on the evaluator's purchase history of clothing. Also, for example, in the example in Figure 2, if it is estimated that the fashion category that the evaluator does not usually purchase is "mode," the selection unit 41 will not exclude the evaluator's evaluation result in the evaluation of target information that shows an outfit belonging to "mode," even if the time from when the target information is displayed until the evaluator selects it is greater than or equal to a predetermined threshold. Furthermore, if it is estimated that the evaluator's usual fashion category is "casual," the selection unit 41 sets a shorter time limit for the evaluator to select a target information item that represents an outfit belonging to "casual" from the time the target information is displayed compared to evaluators who do not usually purchase "casual" items. If the time limit is exceeded, the evaluation result by that evaluator is excluded, and new target information is selected.

[0145] Furthermore, the selection unit 41 may select new target information based on evaluation results excluding the evaluation result of the evaluator if the evaluator's evaluation of the target information satisfies conditions set based on the evaluator's browsing history of the items being evaluated. For example, in the example in Figure 2, the selection unit 41 selects new target information based on evaluation results excluding the evaluation result of the evaluator if the evaluator's evaluation of the target information satisfies conditions set based on the evaluator's browsing history of the outfits. Also, for example, in the example in Figure 2, if it is estimated that the fashion category that the evaluator does not usually browse is "mode," the selection unit 41 will not exclude the evaluation result of the evaluator in the evaluation of target information that represents outfits belonging to "mode," even if the time from when the target information is displayed until the evaluator selects it is greater than or equal to a predetermined threshold. Furthermore, if it is estimated that the fashion category the evaluator usually browses is "casual," the selection unit 41 sets a shorter time limit for the evaluator to select target information that represents outfits belonging to "casual" from the time the target information is displayed compared to evaluators who do not usually browse "casual" items. If the time limit is exceeded, the evaluation result by that evaluator is excluded, and new target information is selected.

[0146] (Regarding reception desk 42) The reception unit 42 receives from the evaluator a first evaluation indicating which of the multiple first target information selected by the selection unit 41 is favorable, and a second evaluation indicating which of the multiple second target information selected by the selection unit 41 is favorable. For example, in the example in Figure 2, the reception unit 42 receives an evaluation (first evaluation) for target information group #1 and stores it in the evaluator information database 32. The reception unit 42 also receives an evaluation (second evaluation) for target information group #2. The reception unit 42 also receives an evaluation (second evaluation) for target information group #3. The reception unit 42 also receives an evaluation (second evaluation) for target information group #4.

[0147] Furthermore, the reception unit 42 may receive evaluations from the evaluator indicating which of the multiple pieces of target information selected by the selection unit 41 is favorable. For example, in the example in Figure 2, the reception unit 42 receives evaluations from the evaluator for the group of target information #1 to #4.

[0148] Furthermore, the reception unit 42 may receive first and second evaluations from multiple evaluators having the corresponding attributes. For example, in the example in Figure 2, the reception unit 42 receives evaluations for target information groups #1 to #4 from multiple evaluators having the corresponding attributes.

[0149] Furthermore, the reception unit 42 may accept first and second evaluations from multiple evaluators who correspond to at least one of the following: gender, age, place of residence, place of employment, preferred categories of items to be evaluated, preferred fashion brands, preferred fashion genres, preferred fashion influencers, level of interest in the items to be evaluated, search trends, browsing trends, purchase trends, purchase amount, and information used as a reference for fashion. For example, in the example in Figure 2, the reception unit 42 accepts evaluations of the target information groups #1 to #4 from multiple evaluators who correspond to at least one of the following attributes: gender, age, preferred fashion categories, and level of interest in fashion.

[0150] Furthermore, the reception unit 42 may receive from the evaluator a first evaluation indicating which of the multiple first target information selected by the selection unit 41 is favorable, and a third evaluation indicating which of the multiple third target information selected by the selection unit 41 is favorable. For example, in the example in Figure 2, the reception unit 42 receives an evaluation from the evaluator for target information to which the additional target "body type: excluding hands" has been added.

[0151] (Regarding Estimation Section 43) The estimation unit 43 estimates the influence of the added item on the evaluation of the evaluation target based on the relative relationship between the first and second evaluations received by the reception unit 42. For example, in the example in Figure 2, the estimation unit 43 estimates the influence of the added item on the evaluation of the code based on the relative relationship of the evaluators' evaluations of the target information group #1 to #4.

[0152] Furthermore, the estimation unit 43 may estimate that the greater the difference between the first evaluation and the second evaluation, the greater the degree of influence that the added item has on the evaluation of the item being evaluated. For example, in the example in Figure 2, the estimation unit 43 estimates that the greater the difference between score Sc1 and score Sc2, the greater the degree of influence that the added item "head" has on the evaluation of the outfit.

[0153] Furthermore, the estimation unit 43 may estimate whether the added item has a positive or negative influence on the evaluation of the item being evaluated, based on the difference between the first evaluation and the second evaluation. For example, in the example in Figure 2, the estimation unit 43 estimates that the added item "head" has a positive influence on the evaluation of the outfit if score Sc2 is higher than score Sc1, and estimates that the added item "head" has a negative influence on the evaluation of the outfit if score Sc2 is lower than score Sc1. To give a specific example, the estimation unit 43 estimates that the higher score Sc2 is compared to score Sc1, the greater the degree of positive influence that the added item "head" has on the evaluation of the outfit, and the lower score Sc2 is compared to score Sc1, the greater the degree of negative influence that the added item "head" has on the evaluation of the outfit.

[0154] Furthermore, the estimation unit 43 may estimate the influence that a part of the added item has on the evaluation of the item being evaluated, based on the relative relationship between the first and third evaluations received by the reception unit 42. For example, in the example in Figure 2, the estimation unit 43 estimates the influence that the added item "Body type: Excluding hands" has on the evaluation of the outfit, based on the evaluation from the evaluator.

[0155] Furthermore, the estimation unit 43 may estimate, based on a first evaluation showing the evaluator's evaluation of the first target information, which is information showing only the target to be evaluated, and a second evaluation showing the evaluator's evaluation of the second target information, which further shows the additional target along with the target to be evaluated, whether the influence of a given user on the evaluation of the target to be evaluated satisfies predetermined conditions. For example, in the example in Figure 2, the estimation unit 43 estimates, based on the relative relationships of the evaluators' evaluations of the target information group #1 to #4, whether the influence of an additional target on the evaluation of the code satisfies predetermined conditions.

[0156] Furthermore, the estimation unit 43 may estimate whether the added item has a positive or negative influence on the user's evaluation of the evaluation item. For example, in the example in Figure 2, the estimation unit 43 estimates whether the added item has a positive or negative influence on the evaluation of the code.

[0157] Furthermore, the estimation unit 43 may estimate, based on the first and second evaluations by evaluators having attributes corresponding to the user, whether the influence of the user on the evaluation target satisfies predetermined conditions. For example, in the example in Figure 2, the estimation unit 43 estimates, based on the evaluations of the target information by groups G1 to G4, to which evaluators having attributes corresponding to the user U1 belong, whether the influence of the added items on the evaluation of the code satisfies predetermined conditions.

[0158] Furthermore, the estimation unit 43 may estimate whether the influence of an additional item on the evaluation of an item by a user satisfies predetermined conditions, based on the first and second evaluations by evaluators corresponding to the user, at least one of the following: gender, age, place of residence, place of employment, preferred categories of items to be evaluated, preferred fashion brands, preferred fashion genres, preferred fashion influencers, level of interest in items to be evaluated, search trends, browsing trends, purchase trends, purchase amount, and information used as a reference for fashion. For example, in the example in Figure 2, the estimation unit 43 estimates whether the influence of an additional item on the evaluation of an outfit satisfies predetermined conditions, based on the evaluations of the target information by user U1 and groups G1 to G4 to which evaluators with corresponding attributes such as gender, age, place of residence, preferred fashion categories, and level of interest in fashion belong.

[0159] (Regarding Section 44) The provisioning unit 44 provides the user with content related to the evaluation target based on the addition target estimated by the estimation unit 43. For example, in the example in Figure 2, the provisioning unit 44 provides user U1 with content related to the outfit based on the impact of the addition target on the outfit.

[0160] Furthermore, the providing unit 44 may provide content to which an additional element indicating at least one of the following is attached to the evaluation target: an image of a person's head, information about the body shape of the person using the evaluation target, the pose of the person using the evaluation target, the manner in which the evaluation target is used, the situation in which the evaluation target is used, and a background image. For example, in the example in Figure 2, the providing unit 44 provides content related to the outfit to user U1 based on the influence that the additional element "head", additional element "body shape", additional element "pose", additional element "style", and additional element "background image" have on the evaluation of the outfit.

[0161] Furthermore, the providing unit 44 may provide content in which an additional element is added to the evaluation target, which is estimated to have an influence on the evaluation of the evaluation target by the user that meets predetermined conditions. For example, in the example in Figure 2, the providing unit 44 provides user U1 with content showing an outfit to which the additional element "head" or the additional element "body shape" is added.

[0162] Furthermore, the providing unit 44 may provide content in which an additional item corresponding to the user's purchase history of the item being evaluated is added to the item being evaluated. For example, in the example in Figure 2, if the additional item "head" is estimated to have a positive influence on the evaluation of the outfit, the providing unit 44 provides content in which an image of the head corresponding to the product image of the clothing purchased by user U1 in an e-commerce service, etc., is added to the outfit.

[0163] Furthermore, the service provider 44 may provide content in which an additional element corresponding to the user's browsing history of the evaluated item is added to the evaluated item. For example, in the example in Figure 2, if the additional element "head" is estimated to have a positive influence on the evaluation of the outfit, the service provider 44 provides content in which an image of the outfit that user U1 reacted positively to in the coordination service, etc., and an image of the head corresponding to that outfit are added to the outfit.

[0164] Furthermore, the provisioning unit 44 may prioritize providing content to which an additional element has been added that is estimated to have a certain degree of influence on the evaluation of the evaluation subject by the user. For example, in the example in Figure 2, the provisioning unit 44 prioritizes providing content related to outfits to which the additional element "head" or the additional element "body type" has been added in e-commerce services and coordination services.

[0165] Furthermore, the providing unit 44 may provide the user with content related to the evaluation target, depending on whether the added object has a positive or negative influence on the user's evaluation of the evaluation target. For example, in the example in Figure 2, the providing unit 44 provides content showing the outfit, depending on whether the added object has a positive or negative influence on the evaluation of the outfit.

[0166] Furthermore, the providing unit 44 may provide content in which an attachment presumed to have a positive influence is attached to the evaluation target, and content in which an attachment presumed to have a negative influence is not attached to the evaluation target. For example, in the example in Figure 2, if the attachment "head" is presumed to have a positive influence on the evaluation of the outfit, the providing unit 44 will provide content showing the outfit to which the attachment "head" is attached. Also, if the attachment "head" is presumed to have a negative influence on the evaluation of the outfit, the providing unit 44 will provide content showing the outfit to which the attachment "head" is not attached.

[0167] Furthermore, the providing unit 44 may prioritize providing content in which an attachment presumed to have a positive influence is attached to the evaluation target over content in which an attachment presumed to have a negative influence is attached to the evaluation target. For example, in the example in Figure 2, the providing unit 44 prioritizes providing content in which an attachment presumed to have a positive influence is attached to the outfit over content in which an attachment presumed to have a negative influence is attached to the outfit.

[0168] (Regarding setting section 45) The setting unit 45 sets that evaluators whose evaluation content for the target information meets predetermined conditions will not be included in the target information presentation. For example, in the example in Figure 2, the setting unit 45 sets evaluators whose evaluation content meets predetermined conditions as excluded.

[0169] (Regarding decision section 46) The decision unit 46 determines whether or not to include evaluators who have been set by the setting unit 45 as subjects to whom the target information will be presented, based on information about those evaluators. For example, in the example in Figure 2, the decision unit 46 determines whether or not to include subjects to whom the target information will be presented, based on information about excluded persons.

[0170] Furthermore, the decision unit 46 may decide whether or not to include an evaluator in the list of those to whom the target information is presented, based on the evaluator's evaluation of the target information. For example, in the example in Figure 2, the decision unit 46 decides whether or not to reinstate an excluded person as an evaluator, based on the excluded person's evaluation of the target information.

[0171] Furthermore, the decision unit 46 may decide whether or not to include an evaluator in the target information presentation based on the evaluator's evaluation of the target information for which the correct answer has been set in advance. For example, in the example in Figure 2, the decision unit 46 presents the target information that is correct and the target information that is incorrect to the excluded person a predetermined number of times, and if the percentage of the target information that is correct is above a predetermined threshold, the excluded person is reinstated as an evaluator.

[0172] Furthermore, the decision unit 46 may decide whether or not to include an evaluator as a subject to present with the subject, based on information about the evaluator and information about the subject. For example, in the example in Figure 2, the decision unit 46 reinstates an excluded person as an evaluator when evaluating subject information related to a fashion category in which the excluded person is an expert.

[0173] Furthermore, the decision unit 46 may determine whether or not to include an evaluator as one of the evaluators to whom the target information is presented, based on information about the evaluator and information about other evaluators who present the target information. For example, in the example in Figure 2, if the decision unit 46 presents the target information to an evaluator whose attributes correspond to those of an excluded person, it reinstates the excluded person as an evaluator.

[0174] (Regarding Learning Section 47) The learning unit 47 trains the model based on the first and second evaluations. For example, in the example in Figure 2, the learning unit 47 trains model #2 using codes with a rating above a predetermined threshold as ground truth data.

[0175] [5. Information Processing Flow] Using Figure 9, the information processing procedure (1) of the information processing device 10 according to the embodiment will be explained. Figure 9 is a flowchart (1) showing an example of the information processing procedure according to the embodiment.

[0176] As shown in Figure 9, the information processing device 10 selects a plurality of first target information and a plurality of second target information to present to the evaluator from among first target information that shows only the target to be evaluated and second target information that shows additional targets along with the target to be evaluated (step S101). Next, the information processing device 10 determines whether or not it has received a first evaluation from the evaluator indicating which of the plurality of first target information is favorable, and a second evaluation from the plurality of second target information indicating which of the plurality of second target information is favorable (step S102). If the first and second evaluations have not been received (step S102; No), the information processing device 10 waits until the first and second evaluations are received.

[0177] On the other hand, if the first and second evaluations are received (step S102; Yes), the information processing device 10 estimates the influence of the added object on the evaluation of the evaluation object based on the relative relationship between the first and second evaluations (step S103), and then terminates the process.

[0178] Next, the information processing procedure (2) of the information processing device 10 according to the embodiment will be described using Figure 10. Figure 10 is a flowchart (2) showing an example of the information processing procedure according to the embodiment.

[0179] As shown in Figure 10, the information processing device 10 determines whether it has received a first evaluation from the evaluator, which is an evaluation of the first target information that shows only the target to be evaluated, and a second evaluation, which is an evaluation of the second target information that shows additional targets along with the target to be evaluated (step S201). If the first and second evaluations have not been received (step S201; No), the information processing device 10 waits until the first and second evaluations are received.

[0180] On the other hand, if the first and second evaluations are received (step S201; Yes), the information processing device 10 estimates, based on the first and second evaluations, additional items whose influence on the evaluation of the evaluation target by a designated user meets predetermined conditions (step S202). Subsequently, the information processing device 10 provides the user with content related to the evaluation target based on the additional items that meet the predetermined conditions (step S203), and terminates the process.

[0181] Next, the information processing procedure (3) of the information processing device 10 according to the embodiment will be described using Figure 11. Figure 11 is a flowchart (3) showing an example of the information processing procedure according to the embodiment.

[0182] As shown in Figure 11, the information processing device 10 selects several pieces of target information to present to the evaluator from among the target information indicating the items to be evaluated (step S301). Next, the information processing device 10 determines whether or not it has received an evaluation from the evaluator indicating which of the multiple pieces of target information is favorable (step S302). If no evaluation has been received (step S302; No), the information processing device 10 waits until an evaluation is received.

[0183] On the other hand, if an evaluation is accepted (Step S302; Yes), the information processing device 10 selects new target information based on the evaluation results, excluding evaluation results from evaluators whose evaluation content for the target information meets predetermined conditions (Step S303), and then terminates the process.

[0184] [6. Variant Example] The above-described embodiment is merely an example, and various modifications and applications are possible.

[0185] [6-1. Regarding the processing method] Of the processes described in the above embodiments, all or part of the processes described as being performed automatically can be performed manually, and conversely, all or part of the processes described as being performed manually can be performed automatically by known methods. In addition, the processing procedures, specific names, and information including various data and parameters shown in the above text and drawings can be arbitrarily changed unless otherwise specified. For example, the various information shown in each figure is not limited to the information shown.

[0186] Furthermore, the components of each illustrated device are functionally conceptual and do not necessarily need to be physically configured as shown. In other words, the specific forms of distribution and integration of each device are not limited to those shown, and all or part of them can be functionally or physically distributed and integrated in any unit according to various loads and usage conditions.

[0187] Furthermore, the embodiments described above can be combined as appropriate, provided that the processing content is not contradictory.

[0188] [7. Effects] As described above, the information processing device 10 according to the embodiment includes a selection unit 41, a reception unit 42, an estimation unit 43, a provision unit 44, a setting unit 45, a determination unit 46, and a learning unit 47. The selection unit 41 selects a plurality of first target information and a plurality of second target information to present to the evaluator from among first target information that shows only the evaluation target and second target information that shows additional targets along with the evaluation target. The selection unit 41 also selects a plurality of first target information and a plurality of second target information to present to the evaluator from among first target information that is an image showing only the clothing and second target information that shows additional targets along with the clothing. Furthermore, the selection unit 41 selects a plurality of first target information and a plurality of second target information to present to the evaluator from among first target information that is an image showing only the combination of a plurality of clothing and second target information that shows additional targets along with the combination of a plurality of clothing. Furthermore, the selection unit 41 selects multiple target information items to present to the evaluator from among the target information items that represent the items to be evaluated.The selection unit 41 then selects new target information based on the evaluation results obtained by the evaluator, excluding evaluation results from evaluators whose evaluation content for the target information satisfies predetermined conditions.The selection unit 41 also selects multiple first target information items to present to the evaluator from among the first target information items that represent combinations of items to be evaluated identified using a model trained to determine whether or not the combination of items to be evaluated satisfies predetermined conditions, and selects multiple second target information items to present to the evaluator from among the second target information items that represent combinations of items to be evaluated identified using the same model.The reception unit 42 receives from the evaluator a first evaluation indicating which of the multiple first target information items selected by the selection unit 41 is favorable, and a second evaluation indicating which of the multiple second target information items selected by the selection unit 41 is favorable.The reception unit 42 also receives from the evaluator an evaluation indicating which of the multiple target information items selected by the selection unit 41 is favorable. The estimation unit 43 estimates the influence of the added object on the evaluation of the evaluation target based on the relative relationship between the first evaluation and the second evaluation received by the reception unit 42.Furthermore, the estimation unit 43 estimates additional items whose influence on the evaluation of the evaluation target by a given user meets predetermined conditions, based on a first evaluation showing the evaluator's evaluation of the first target information, which is information that shows only the evaluation target, and a second evaluation showing the evaluator's evaluation of the second target information, which shows additional items along with the evaluation target. The provision unit 44 provides content related to the evaluation target to the user based on the additional items estimated by the estimation unit 43. The setting unit 45 sets evaluators whose evaluation content of the target information meets predetermined conditions not to be included as targets for which the target information is presented. The decision unit 46 decides whether or not to include evaluators who have been set by the setting unit 45 not to be included as targets for which the target information is presented, based on information about those evaluators. The learning unit 47 learns the model based on the first evaluation and the second evaluation.

[0189] As a result, the information processing device 10 according to the embodiment can understand the influence that additional elements such as head shape and body shape have on the evaluation of an outfit, not just the compatibility of the clothing that makes up the outfit. Therefore, it can understand the influence that elements added to the item being evaluated have on the evaluation of the item being evaluated. Furthermore, the information processing device 10 according to the embodiment can prioritize providing the user with content to which additional elements that are presumed to have a positive influence on the evaluation of the outfit have been added, and if such additional elements are not added, it can provide content to which such additional elements have been added. It can understand the influence that elements such as head shape and body shape have on the evaluation of an outfit. In other words, the information processing device 10 according to the embodiment can provide content related to the item being evaluated according to the influence that elements added to the item being evaluated have on the evaluation of the item being evaluated. Furthermore, the information processing device 10 according to the embodiment can exclude evaluation results from evaluators presumed to be unserious, set appropriate ratings based on evaluation results from other serious evaluators, and appropriately select the target information to present to the evaluator. Therefore, it can select the information to present to the evaluator according to the content of the evaluator's evaluation.

[0190] Furthermore, in the information processing device 10 according to the embodiment, for example, the selection unit 41 selects a plurality of second target information to present to the evaluator from among the second target information which indicates at least one of the following as an additional target: an image of a person's head, information about the body shape of the person using the object to be evaluated, the pose of the person using the object to be evaluated, the manner in which the object to be evaluated is used, the situation in which the object to be evaluated is used, and a background image.

[0191] As a result, the information processing device 10 according to the embodiment can understand the influence that various added elements have on the evaluation of the evaluation target, thereby improving convenience.

[0192] Furthermore, in the information processing device 10 according to the embodiment, for example, the selection unit 41 selects a plurality of first target information to present to the evaluator from among the first target information based on the rating of the first target information calculated based on the first evaluation, and selects a plurality of second target information to present to the evaluator from among the second target information based on the rating of the second target information calculated based on the second evaluation. The selection unit 41 selects a plurality of first target information from among the first target information in which the difference in ratings is within a predetermined range, and selects a plurality of second target information from among the second target information in which the difference in ratings is within a predetermined range. The selection unit 41 selects a plurality of first target information to present to the evaluator from among the first target information based on the number of times the first target information has been evaluated by the evaluator, and selects a plurality of second target information to present to the evaluator from among the second target information based on the number of times the second target information has been evaluated by the evaluator.

[0193] As a result, the information processing device 10 according to this embodiment can improve the accuracy of rating the target information and increase the amount of information obtained by evaluating the target information, thereby enabling efficient evaluation of the target information.

[0194] Furthermore, in the information processing device 10 according to the embodiment, for example, the reception unit 42 receives first and second evaluations from multiple evaluators having corresponding attributes. The reception unit 42 also receives first and second evaluations from multiple evaluators who have at least one of the following characteristics: gender, age, place of residence, place of employment, preferred category of items to be evaluated, preferred fashion brands, preferred fashion genres, preferred fashion influencers, level of interest in the items to be evaluated, search trends, browsing trends, purchase trends, purchase amount, and information used as a reference for fashion.

[0195] As a result, the information processing device 10 according to this embodiment can receive evaluations from evaluators having predetermined attributes, and users having predetermined attributes can understand the impact that the added object has on the evaluation of the object being evaluated.

[0196] Furthermore, in the information processing device 10 according to the embodiment, for example, the estimation unit 43 estimates that the greater the difference between the first evaluation and the second evaluation, the greater the degree of influence that the added object has on the evaluation of the evaluation target. Also, based on the difference between the first evaluation and the second evaluation, the estimation unit 43 estimates whether the added object has a positive or negative influence on the evaluation of the evaluation target.

[0197] As a result, the information processing device 10 according to this embodiment can grasp the degree to which the added object has an influence on the evaluation of the evaluation target, and what kind of influence that influence is, thus enabling accurate understanding of the influence of the added object on the evaluation of the evaluation target.

[0198] Furthermore, in the information processing device 10 according to the embodiment, for example, the selection unit 41 selects a plurality of first target information and a plurality of third target information to present to the evaluator from among first target information that shows only the target to be evaluated and third target information that shows a portion of the additional objects that are estimated to have an impact on the evaluation of the target to be evaluated, together with the target to be evaluated.The reception unit 42 then receives from the evaluator a first evaluation indicating which of the plurality of first target information selected by the selection unit 41 is favorable, and a third evaluation indicating which of the plurality of third target information selected by the selection unit 41 is favorable.The estimation unit 43 then estimates the impact that a portion of the additional objects has on the evaluation of the target to be evaluated, based on the relative relationship between the first evaluation and the third evaluation received by the reception unit 42.

[0199] As a result, the information processing device 10 according to this embodiment can subdivide the added elements that affect the evaluation of the code and estimate their influence, so it can accurately determine which part of the added elements has an influence.

[0200] Furthermore, in the information processing device 10 according to the embodiment, for example, the providing unit 44 provides content to which an additional object is attached to the evaluation object, which includes an additional object showing at least one of the following: an image of a person's head, information about the body shape of a person using the evaluation object, the pose of the person using the evaluation object, the manner in which the evaluation object is used, the situation in which the evaluation object is used, and a background image.

[0201] As a result, the information processing device 10 according to the embodiment can provide content related to the evaluation target to which various additional elements have been added, depending on whether or not they affect the evaluation of the evaluation target, thereby improving convenience.

[0202] Furthermore, in the information processing device 10 according to the embodiment, for example, the providing unit 44 provides the content relating to the evaluation target, taking into account additional objects that are estimated to satisfy predetermined conditions for influence on the evaluation of the evaluation target by the user. The providing unit 44 also provides content to the evaluation target to which additional objects that are estimated to satisfy predetermined conditions for influence on the evaluation of the evaluation target by the user have been added. The providing unit 44 also provides content to the evaluation target to which additional objects corresponding to the user's purchase history of the evaluation target have been added. The providing unit 44 also provides content to the evaluation target to which additional objects corresponding to the user's browsing history of the evaluation target have been added.

[0203] As a result, the information processing device 10 according to this embodiment can provide content related to an evaluation target to which additional elements corresponding to the user have been added, thereby providing content with a high appeal.

[0204] Furthermore, in the information processing device 10 according to the embodiment, for example, the provisioning unit 44 preferentially provides content to which an added object has been added to the evaluation target, and which is estimated to have an influence on the evaluation of the evaluation target by the user that satisfies predetermined conditions.

[0205] As a result, the information processing device 10 according to the embodiment can prioritize providing content related to an evaluation target to which an additional object has been added that satisfies predetermined conditions for influencing the evaluation of the evaluation target, thus prioritizing the provision of content with a high appeal effect.

[0206] Furthermore, in the information processing device 10 according to the embodiment, for example, the estimation unit 43 estimates whether the added object has a positive or negative influence on the user's evaluation of the evaluation target. The provision unit 44 then provides content related to the evaluation target to the user according to whether the added object has a positive or negative influence on the user's evaluation of the evaluation target. The provision unit 44 also provides content to the evaluation target in which the added object estimated to have a positive influence has been added, and provides content to the evaluation target in which the added object estimated to have a negative influence has not been added. In addition, the provision unit 44 provides content to the evaluation target in which the added object estimated to have a positive influence has been added, with priority given to providing content to the evaluation target in which the added object estimated to have a negative influence has been added.

[0207] As a result, the information processing device 10 according to the embodiment can provide content related to the evaluation target to which the added object has been attached, depending on whether it has a positive or negative influence, and thus can provide content with a high appeal.

[0208] Furthermore, in the information processing device 10 according to the embodiment, for example, the estimation unit 43 estimates additional items whose influence on the evaluation of the evaluation subject by the user is satisfied with predetermined conditions, based on first and second evaluations by evaluators having attributes corresponding to the user. In addition, the estimation unit 43 estimates additional items whose influence on the evaluation of the evaluation subject by the user is satisfied with predetermined conditions, based on first and second evaluations by evaluators having attributes corresponding to the user, at least one of the following: gender, age, place of residence, place of employment, preferred categories of evaluation subjects, preferred fashion brands, preferred fashion genres, preferred fashion influencers, level of interest in the evaluation subject, search trends, browsing trends, purchase trends, purchase amount, and information used as a reference for fashion.

[0209] As a result, the information processing device 10 according to the embodiment can estimate the impact of the added object based on evaluations by evaluators having attributes corresponding to the user, and provide content, thereby providing content appropriate to the user's attributes.

[0210] Furthermore, in the information processing device 10 according to the embodiment, for example, if the evaluator evaluates dummy information presented with the target information, or dummy information presented in place of the target information, as favorable, the selection unit 41 selects new target information based on the evaluation results excluding the evaluator's evaluation result. Also, if the evaluator evaluates any of the same target information as favorable, the selection unit 41 selects new target information based on the evaluation results excluding the evaluator's evaluation result. Also, if the evaluator evaluates target information that has been set as correct in advance as unfavorable, the selection unit 41 selects new target information based on the evaluation results excluding the evaluator's evaluation result. Also, if the time required by the evaluator to evaluate the target information satisfies predetermined conditions, the selection unit 41 selects new target information based on the evaluation results excluding the evaluator's evaluation result. Also, if the position on the screen of the terminal device used by the evaluator where the evaluator performed an operation related to the evaluation of the target information satisfies predetermined conditions, the selection unit 41 selects new target information based on the evaluation results excluding the evaluator's evaluation result. Furthermore, if the evaluator performs an operation on the terminal device used by the evaluator to display a screen different from the screen used to evaluate the target information, the selection unit 41 selects new target information based on the evaluation results excluding the evaluator's evaluation results. Also, if the evaluator's evaluation of the target information satisfies the conditions set based on information about the evaluator, the selection unit 41 selects new target information based on the evaluation results excluding the evaluator's evaluation results. Furthermore, if the evaluator's evaluation of the target information satisfies the conditions set based on the evaluator's purchase history of the item being evaluated, the selection unit 41 selects new target information based on the evaluation results excluding the evaluator's evaluation results.

[0211] As a result, the information processing device 10 according to the embodiment can appropriately select the target information to present to the evaluator based on the evaluation results of other serious evaluators, by excluding evaluation results from evaluators who are presumed to be unserious through various methods.

[0212] Furthermore, in the information processing device 10 according to the embodiment, for example, the decision unit 46 determines whether or not to include an evaluator as a target to present the target information, based on the evaluator's evaluation of the target information. The decision unit 46 also determines whether or not to include an evaluator as a target to present the target information, based on the evaluator's evaluation of the target information for which the correct answer has been set in advance. The decision unit 46 also determines whether or not to include an evaluator as a target to present the target information, based on information about the evaluator and information about the target information. The decision unit 46 also determines whether or not to include an evaluator as a target to present the target information, based on information about the evaluator and information about other evaluators who present the target information.

[0213] As a result, the information processing device 10 according to this embodiment can reinstate an evaluator if the excluded person meets predetermined conditions, thereby allowing evaluators who were excluded for inappropriate reasons to be reinstated.

[0214] [8. Hardware Configuration] Furthermore, the information processing device 10 according to each embodiment described above can be implemented by a computer 1000 having a configuration such as that shown in Figure 12. The following explanation will use the information processing device 10 as an example. Figure 12 is a hardware configuration diagram showing an example of a computer that implements the functions of the information processing device 10. The computer 1000 has a CPU 1100, RAM 1200, ROM 1300, HDD 1400, communication interface (I / F) 1500, input / output interface (I / F) 1600, and media interface (I / F) 1700.

[0215] The CPU 1100 operates based on programs stored in the ROM 1300 or HDD 1400, and controls various parts. The ROM 1300 stores boot programs executed by the CPU 1100 when the computer 1000 starts up, as well as programs that depend on the computer 1000's hardware.

[0216] The HDD 1400 stores programs executed by the CPU 1100, as well as data used by such programs. The communication interface 1500 receives data from other devices via the communication network 500 (corresponding to network N in this embodiment) and sends it to the CPU 1100, and also transmits data generated by the CPU 1100 to other devices via the communication network 500.

[0217] The CPU 1100 controls output devices such as displays and printers, and input devices such as keyboards and mice, via the input / output interface 1600. The CPU 1100 acquires data from input devices via the input / output interface 1600. The CPU 1100 also outputs the data it generates to output devices via the input / output interface 1600.

[0218] The media interface 1700 reads a program or data stored in the recording medium 1800 and provides it to the CPU 1100 via the RAM 1200. The CPU 1100 loads the program from the recording medium 1800 onto the RAM 1200 via the media interface 1700 and executes the loaded program. The recording medium 1800 is, for example, an optical recording medium such as a DVD (Digital Versatile Disc) or PD (Phase Change Rewritable Disk), a magneto-optical recording medium such as an MO (Magneto-Optical disk), a tape medium, a magnetic recording medium, or a semiconductor memory.

[0219] For example, when computer 1000 functions as information processing device 10, the CPU 1100 of computer 1000 realizes the functions of control unit 40 by executing programs loaded on RAM 1200. The HDD 1400 stores the data in the storage device of information processing device 10. The CPU 1100 of computer 1000 reads and executes these programs from the recording medium 1800, but as another example, these programs may be obtained from other devices via a predetermined communication network.

[0220] [9. Other] Although some embodiments of the present invention have been described in detail above with reference to the drawings, these are illustrative examples, and the present invention can be implemented in various other forms with modifications and improvements based on the knowledge of those skilled in the art, starting with the embodiments described in the disclosure section of the invention.

[0221] Furthermore, the configuration of the aforementioned information processing device 10 can be flexibly changed, for example, by calling external platforms, etc., via APIs (Application Programming Interfaces) or network computing, depending on the function.

[0222] Furthermore, the term "part" in the claims can be replaced with "means," "circuit," etc. For example, "selection part" can be replaced with "selection means" or "selection circuit." [Explanation of symbols]

[0223] 10 Information Processing Devices 20 Communications Department 30 Storage section 31 Target Information Database 32. Evaluator Information Database 33. User Information Database 40 Control Unit 41 Selection Section 42 Reception Department 43 Estimation part 44 Providing Department 45 Settings section 46 Decision Section 47. Learning Department 100 User Terminals 200 evaluator terminals

Claims

1. An estimation unit estimates additional targets whose influence on the evaluation of the evaluation target by a predetermined user satisfies predetermined conditions, based on a first evaluation showing the evaluator's evaluation of first target information, which is information that shows only the evaluation target, and a second evaluation showing the evaluator's evaluation of second target information, which further shows additional targets along with the evaluation target. Based on the additional target estimated by the estimation unit, the provision unit provides content related to the evaluation target to the user. An information processing device characterized by having the following features.

2. The aforementioned supply unit is, The content provided includes an additional element attached to the evaluation target that shows at least one of the following: an image of a person's head, information about the body shape of the person using the evaluation target, the pose of the person using the evaluation target, the manner in which the evaluation target is used, the situation in which the evaluation target is used, and a background image. The information processing apparatus according to feature 1.

3. The aforementioned supply unit is, The content provided concerns an evaluation target, taking into account additional items whose influence on the evaluation by the user is estimated to meet predetermined conditions. The information processing apparatus according to feature 1.

4. The aforementioned supply unit is, The content provided is an additional element that is attached to the evaluation target, and whose influence on the evaluation of the evaluation target by the user is estimated to meet predetermined conditions. The information processing apparatus according to feature 1.

5. The aforementioned supply unit is, The content to which the purchase history of the item to be evaluated by the user is added is provided. The information processing apparatus according to feature 4.

6. The aforementioned supply unit is, The content to which the user's browsing history of the subject to evaluation is added is provided. The information processing apparatus according to feature 4.

7. The aforementioned supply unit is, The content attached to the evaluation target will be preferentially provided to users whose influence on the evaluation of the evaluation target is estimated to meet the predetermined conditions. The information processing apparatus according to feature 1.

8. The estimation unit, We estimate whether the added item has a positive or negative influence on the user's evaluation of the item being evaluated. The aforementioned supply unit is, Depending on whether the added content has a positive or negative influence on the user's evaluation of the evaluation subject, content related to the evaluation subject will be provided to the user. The information processing apparatus according to feature 1.

9. The aforementioned supply unit is, Content is provided that includes elements estimated to have a positive influence, and content is provided that does not include elements estimated to have a negative influence. The information processing apparatus according to feature 8.

10. The aforementioned supply unit is, Content with attachments estimated to have a positive influence will be given priority over content with attachments estimated to have a negative influence. The information processing apparatus according to feature 7.

11. The estimation unit, Based on the first and second evaluations by evaluators with attributes corresponding to the aforementioned user, the system estimates additional targets whose influence on the evaluation of the evaluation target by the aforementioned user satisfies predetermined conditions. The information processing apparatus according to feature 1.

12. The estimation unit, Based on the first and second evaluations by evaluators corresponding to the user, at least one of the following factors—gender, age, place of residence, place of employment, preferred categories of items to evaluate, preferred fashion brands, preferred fashion genres, preferred fashion influencers, level of interest in items to evaluate, search trends, browsing trends, purchase trends, purchase amount, and information used as reference for fashion—the system estimates additional items whose influence on the evaluation of items by the user meets predetermined conditions. The information processing apparatus according to feature 11.

13. A method of information processing performed by a computer, An estimation step is performed to estimate additional targets whose influence on the evaluation of the evaluation target by a predetermined user satisfies predetermined conditions, based on a first evaluation which shows the evaluator's evaluation of the first target information which is information that shows only the evaluation target, and a second evaluation which shows the evaluator's evaluation of the second target information which shows additional targets along with the evaluation target. Based on the additional target estimated by the estimation process, a provision process is performed to provide content related to the evaluation target to the user. An information processing method characterized by including

14. An estimation procedure for estimating additional targets whose influence on the evaluation of the evaluation target by a specified user satisfies predetermined conditions, based on a first evaluation showing the evaluator's evaluation of the first target information, which is information that shows only the evaluation target, and a second evaluation showing the evaluator's evaluation of the second target information, which shows additional targets along with the evaluation target. A provision procedure for providing content related to the evaluation target to the user, based on the additional target estimated by the estimation procedure described above. An information processing program characterized by causing a computer to execute it.

Images